Method and system for automatic crystallographic axis alignment

Abstract

Methods and systems for automatic alignment of crystallographic axes. Automatic alignment of a sample's crystallographic axes with a charged particle beam is achieved based on a diffraction pattern of the sample. A region corresponding to a Laue circle is segmented using a trained network. The sample is aligned with the charged particle beam by tilting the sample with a crystallographic axis tilt determined based on the segmented region.

Description

Technical Field

[0001] This specification generally relates to methods and systems for aligning a sample with an incident beam, and more specifically, to methods and systems for automatically aligning the zone axis of a crystalline sample with a beam of charged particles. Background Technology

[0002] For high-resolution charged particle beam microscopy, the charged particle beam must be aligned with the zone axis of the crystalline sample in order to image the sample with high precision. If the zone axis of the sample is not aligned, for example when the zone axis is not parallel to the incident beam orientation, the measurement of nanoscale features on the sample may be inaccurate. The process of aligning the sample's crystal structure with the incident beam is called zone axis alignment.

[0003] One method for zone axis alignment is based on diffraction patterns formed by parallel beams. When a beam of parallel charged particles passes through a thin crystalline sample, the charged particles interfere with each other and form a diffraction pattern on the back focal plane of an objective lens located below the sample. The diffraction pattern consists of multiple bright diffraction points. Diffraction points belonging to the zero-order Laue circle can be identified, and zone axis misalignment can be determined based on the position of the center of the zero-order Laue circle and the center of the direct beam. However, the applicant recognizes that under certain conditions, diffraction points of the zero-order Laue circle are not easily identifiable. As an example, when the sample is curved or bent, diffraction points with different zone axis orientations can be mixed in the diffraction pattern. As another example, under converging beam illumination, diffraction points become disks and can be elongated and overlap with each other and / or the direct beam. Summary of the Invention

[0004] In one embodiment, a method can be used to align the zone axis of a sample, the method comprising directing a beam of charged particles toward the sample; acquiring a diffraction pattern of the sample; segmenting regions of the diffraction pattern corresponding to Laue circles using a trained network; determining a zone axis tilt based on the segmented regions; and tilting the sample based on the determined zone axis tilt. In this manner, the zone axis of a curved sample can be automatically aligned based on a diffraction pattern acquired by a converging beam.

[0005] It should be understood that the above description of the invention is provided to introduce some concepts further described in the detailed embodiments in a simplified form. This is not intended to identify key or essential features of the claimed subject matter, the scope of which is uniquely defined by the claims following the detailed embodiments. Furthermore, the claimed subject matter is not limited to embodiments that address any shortcomings pointed out in the foregoing or any part of this disclosure. Attached Figure Description

[0006] Figure 1 A system for imaging a sample is shown according to some embodiments.

[0007] Figure 2A and Figure 2B The relationship between the incident beam and the sample is shown.

[0008] Figure 3 This is a high-order flowchart of a method for zone axis alignment.

[0009] Figure 4 The coordinate system used to adjust the orientation of the sample is shown.

[0010] Figure 5A This is an exemplary diffraction pattern obtained from a converging beam.

[0011] Figure 5B The output of the trained network is shown.

[0012] Figure 5C The locations of the Laue center and the center of the direct beam are shown.

[0013] Figure 5D It is the diffraction pattern of a sample with its zone axis aligned with the incident beam.

[0014] Figure 6 This is a flowchart of the method used to train the network.

[0015] Figure 7A An exemplary annotated diffraction pattern is shown.

[0016] Figure 7B This shows an example output of the network.

[0017] Throughout the various views in the accompanying drawings, the same reference numerals refer to the corresponding parts. Detailed Implementation

[0018] The following description relates to systems and methods for aligning the zone axis of a sample with a beam of charged particles. For example, to perform high-resolution imaging, a system may be used... Figure 1 For high-resolution scanning transmission electron microscopy (STEM) imaging systems, the zone axis of the crystalline sample must be parallel to the incident beam. In STEM, in response to charged particles irradiating one side of the sample, a detector detects charged particles on the other side of the sample, including transmitted charged particles, scattered electrons, and second electrons.

[0019] Samples imaged in transmission mode may bend due to reduced sample thickness or poor mechanical support. As an example, samples may be as thin as 2 μm. 2 Within a small region, there is a curvature greater than 5 degrees. For example... Figure 2AAs shown, because sample 202 is not flat under the illumination of parallel beam 201, diffraction points from different crystal orientations can be mixed in the diffraction pattern. Therefore, it is difficult to align the charged particle beam with one of the multiple crystal orientations based on the diffraction pattern. Conversely, due to its small beam profile on the sample surface, such as Figure 2B As shown, the diffraction pattern acquired by the converging beam 203 is formed from a sample with uniform crystal orientation. However, the diffraction pattern formed by the converging beam is more complex than that formed by the parallel beam. Unlike the points resembling high-intensity diffraction points in the parallel diffraction pattern, the diffraction points formed by the converging beam diffraction pattern are disk-shaped, elongated, and overlap with each other and with the straight beam.

[0020] Figure 3 This paper illustrates a method for zone axis alignment based on a diffraction pattern. Diffraction patterns can be acquired using either a converging or parallel beam. A trained network is used to segment the diffraction pattern into a first region corresponding to the Laue circle and a second region corresponding to the direct beam. A quality factor representing the segmentation quality is generated based on the output of the trained network. In one instance, the quality factor is generated based on the shape of the segmented region corresponding to the Laue circle. In another instance, the quality factor is generated based on the relative positions of the two segmented regions. Zone axis tilt can be determined based on the positions of the segmented regions. For example, the angles of zone axis tilt in two orthogonal tilt directions are derived based on the relative positions of the centers of the first and second segmented regions. Sample orientation can be adjusted based on the zone axis tilt to align the zone axis of the sample with the incident beam. Sample orientation can be further adjusted until the zone axis tilt converges to a threshold zone axis tilt. Figure 4 The coordinate system used to adjust the orientation of the sample is shown. Figures 5A to 5B An exemplary result of zone axis alignment is shown.

[0021] exist Figure 6 The diagram illustrates a method for generating the trained network. The network can be trained using multiple diffraction patterns acquired by tilting a reference sample at multiple tilt angles. The diffraction patterns of the reference sample can be acquired under the same beam conditions as the sample used for imaging. The Laue circle and the region of the direct beam in the multiple diffraction patterns are automatically annotated based on the known tilt angles. The Laue circle and the region of the direct beam can each be a circle. Multiple diffraction patterns and annotated diffraction patterns are used to train the network. The network output corresponds to two segmented regions in the diffraction patterns: the Laue circle and the directional beam. The network parameters can be adjusted based on the similarity between the segmented Laue circle and the annotated Laue circle. Figures 7A to 7B Exemplary annotated diffraction patterns and network outputs are shown.

[0022] See also Figure 1This diagram illustrates a STEM system 100 according to an embodiment of the present disclosure. The STEM system 100 includes an electron source 10 that emits charged particles, such as an electron beam 11, toward a focusing column 12. The electron beam can generate high-energy electrons, i.e., electrons having typical energies between about 10 keV and 1,000 keV. In some embodiments, the focusing column 12 may include one or more of a condenser lens 121, an aperture 122, a scanning coil 123, and an upper objective lens 124. The focusing column 12 focuses electrons from the electron source 10 onto a small point on a sample 14. Different locations on the sample can be scanned by adjusting the direction of the electron beam via the scanning coil 123. For example, by operating the scanning coil 123, the incident beam 112 (as shown in dashed lines) can be shifted or scanned to focus on different locations on the sample 14. The sample 14 may be thin enough not to obstruct the transmission of most electrons in the electron beam 11.

[0023] The main axis 110 of the imaging system can be the central axis of the electron beam emitted from the electron source 10. The main axis 110 can also be the central axis of the condenser lens 121. When the incident beam is not deflected or scanned (i.e., incident beam 112), the incident beam can be focused at the position where the main axis 110 intersects with the sample 14.

[0024] Sample 14 can be held by sample holder 13. Sample holder 13 can adjust sample orientation by tilting and / or translating the sample. As an example, Figure 4 The coordinate system used to adjust the sample orientation is shown. Figure 4 In this imaging system, the incident beam 112 can be focused onto the sample 14 along the main axis 110. The z-axis can be parallel to the main axis 110. The xy-plane can be a plane perpendicular to the z-axis. The sample 14 can be tilted relative to the main axis 110 by rotation about the x-axis or about the y-axis. For example, the rotation direction about the x-axis can be an α tilt direction 1001, and the rotation direction about the y-axis can be a β tilt direction 1002. The sample holder can also translate or offset the sample 14 along any one of the x-axis, y-axis, and z-axis. In some embodiments, the sample 14 can be rotated about the z-axis.

[0025] Return to Figure 1 Electrons 101 passing through sample 14 can enter projector 116. In one embodiment, projector 116 may be a portion separate from the focusing column. In another embodiment, projector 116 may be an extension of the lens field from the lens in focusing column 12.

[0026] The projector 116 can be adjusted via controller 30 so that direct electrons passing through the sample strike the disk-shaped bright-field detector 115, while diffracted or scattered electrons, more strongly deflected by the sample, are detected by one or more of the high-angle annular dark-field (HAADF) detector 18 and annular dark-field (ADF) detector 19. Signals from the HAADF and ADF detectors can be amplified by amplifiers 20 and 21, respectively. The signal from the bright-field detector 115 can be amplified by amplifier 22. Signals from amplifiers 20, 21, and 22 can be sent to image processor 24, which can form an image of the sample 14 based on the detected electrons. The HAADF detector 18, ADF detector 19, and bright-field detector 115 can be scintillator-photomultiplier detectors, solid-state PIN detectors, or metal plates. The STEM system 100 can simultaneously detect signals from one or more of the ADF, ADF, and HAADF detectors.

[0027] The axial region of sample 14 can be aligned with the incident beam 112 based on the diffraction pattern of sample 14 obtained when the sample is irradiated with the incident beam 112. In one embodiment, the diffraction pattern can be obtained via camera 142 by acquiring the diffraction pattern formed on fluorescent screen 141. During zone axis alignment, fluorescent screen 141 can be inserted between projector 116 and bright-field detector 115. For example, fluorescent screen 141 can be positioned between HAADF detector 18 and ADF detector 19. The HAADF detector can be retracted to acquire the diffraction pattern. In another embodiment, the diffraction pattern on the fluorescent screen can be acquired via camera 143 positioned downstream of bright-field detector 115. Camera 143 can be a CCD or CMOS camera. In some embodiments, the diffraction pattern can be acquired via a pixelated detector. The pixelated detector can also be used to detect one or more of the bright-field, ADF, and HAADF images. The acquired diffraction pattern can be sent to controller 30 for determining zone axis tilt.

[0028] The controller 30 can manually or automatically control the operation of the STEM system 100 in response to operator instructions or according to computer-readable instructions stored in non-transitory memory 32. The controller 30 can be configured to execute computer-readable instructions and control the various components of the STEM system 100 to implement any of the methods described herein. For example, the controller can adjust the beam position on the sample by operating the scanning coil 123. The controller can adjust the profile of the incident beam by adjusting one or more apertures and / or lenses in the focusing column 12. The controller can adjust the sample orientation relative to the incident beam by tilting the sample holder 13. The controller can offset the sample relative to the incident beam by translating the sample holder 13. The controller 30 can be further coupled to a display 31 to display notifications and / or images of the sample. The controller 30 can receive user input from a user input device 33. The user input device 33 may include a keyboard, mouse, or touchscreen.

[0029] Although STEM systems are described with the aid of examples, it should be understood that this technique can be used to align the zone axis with the collimated incident beam. This technique may also be useful when applied to sample alignment in other charged particle beam microscopy systems, such as transmission electron microscopy (TEM) systems, scanning electron microscopy (SEM) systems, and dual-beam microscopy systems. The current discussion of STEM imaging is provided only as an example of a suitable imaging modality.

[0030] Figure 3 Showing for use in, for example Figure 1 A method 300 for aligning the zone axis of a sample with the incident beam in an imaging system of a STEM system. The sample can be curved or bent. In one example, the sample is at 2 μm. 2 The region has a curvature equal to or greater than 0.5 degrees. In another example, the sample is at 2 μm. 2 The region has a curvature equal to or greater than 5 degrees. Based on the converging beam diffraction pattern, the zone axis at a specific sample location can be aligned with the incident beam (e.g., Figure 1 The spindle 110 is automatically aligned.

[0031] At step 302, check the condition of the imaging system. Checking the system condition may include checking one or more of the following: whether the system is operable, whether a suitable sample has been inserted, and whether the required system settings are in place. Operating systems may include, but are not limited to, the column opening valve, the operating electron source, and the functional system vacuum. System settings include, but are not limited to, the required aperture size, condenser lens current, beam position, camera length, electron potential, and beam current. Step 302 also includes acquiring a low-resolution, large field-of-view (FOV) image of the sample. The sample image may be a STEM image. The sample image can be used to locate a region of interest (ROI) for high-resolution imaging. For example, acquiring a large FOV STEM image at 5,000X magnification.

[0032] At position 304, an initial sample image of the ROI is acquired. This initial sample image has higher resolution and a smaller FOV than the sample image acquired at position 302. The initial sample image can be acquired using a HAADF detector.

[0033] At 306, the diffraction pattern of the sample is acquired. The diffraction pattern can be acquired by focusing a beam of charged particles at a single point within the region of interest (ROI). For example, the diffraction pattern is acquired by focusing the beam of charged particles at the center of the ROI and receiving the charged particles with a detector in transmission mode. During diffraction pattern acquisition, the HAADF detector retracts. This can be utilized... Figure 1 Camera 143 acquires diffraction patterns.

[0034] At position 308, the trained network receives the diffraction pattern and outputs two segmented regions of the diffraction pattern. Figure 5A An exemplary diffraction pattern formed by a converging beam of charged particles is shown. The direct beam 503 is a bright circle produced by the incident beam that passes through the sample and strikes the detector directly without scattering. The diffraction point 504 of the Laue circle overlaps with the direct beam. Figure 5B The output of the trained network is shown. The output of the trained network is an image with the same size (or pixels) as the diffraction pattern. The output image contains two segmented regions 501 and 502. The first segmented region 501 corresponds to the Laue circle, and the second segmented region 502 corresponds to the direct beam.

[0035] At 310, a quality factor representing the quality of the trained network output (or the segmentation quality of the trained network) is compared to a threshold quality factor. In one instance, the quality factor is determined based on the shape of the segmented regions corresponding to the Laue circles. If the shape of the segmented Laue circles in the diffraction pattern is closer to a circle, the quality factor is higher. In another instance, if the two segmented regions do not intersect, the quality factor is lower. In yet another instance, the quality factor can be determined based on the degree of overlap between the two segmented regions. The greater the overlap, the higher the quality factor. If the quality factor is greater than a predetermined threshold quality factor, method 300 proceeds to step 310; otherwise, the operator can be notified at 322.

[0036] At step 326, the diffraction pattern can optionally be used to update the parameters of the trained network at step 326. Step 326 also includes determining the validity of the diffraction pattern. Amorphous materials may result in invalid diffraction patterns, in which case zone axis alignment cannot be performed. If the diffraction pattern is valid, it can be annotated and used to update and retrain the trained network.

[0037] At position 312, the zone axis tilt is determined based on the network output. The zone axis tilt comprises a tilt angle relative to the two orthogonal axes (e.g., relative to...). Figure 4 The α and β tilts of the x and y axes can be used to determine the zone axis tilt based on the positions of the centers of the two segmented regions in the trained network output. In one instance, the center of each segmented region can be the geometric center of the segmented region. In another instance, one or more segmented regions can be fitted with a circle. The center of the segmented region is the center of the fitted circle. In yet another instance, the trained network is used to segment only the regions corresponding to the Laue circle, without segmenting the regions corresponding to the direct beam. The direct beam position can be at a fixed, known location, in which case only the segmented regions of the Laue circle are needed.

[0038] Figure 5C The center 505 of the segmented region 502 corresponding to the direct beam is shown, and the center 505 corresponding to the direct beam is shown. Figure 5B The network output of the Laue circle is divided into regions 501 and 506. Relative to... Figure 5A The diffraction pattern shows these centers. The distance 507 between the centers of the two segmented regions increases with the degree of misalignment between the zone axis and the incident beam axis. The zone axis tilt increases with distance 507. At the two orthogonal axes (e.g., Figure 4 The distance 507 on each of the x-axis and y-axis increases with the increase of the tilt angle of the zone axis about the axis.

[0039] At 314, the zone axis tilt determined at 312 is compared with a threshold zone axis tilt to assess the degree of misalignment. In one example, the zone axis tilts in the α and β directions can be compared with threshold α and β region tilt thresholds, respectively. The α and β zone axis tilt thresholds can be 0.5 degrees. If the zone axis tilt is less than the threshold zone axis tilt, the zone axis is aligned, and method 300 proceeds to 324 to obtain a high-resolution sample image. Otherwise, the sample is tilted at 316 to align the sample's zone axis with the incident beam.

[0040] In another example, instead of tilting the zone axis, the degree of overlap between the direct beam and the Laue circle is compared to a threshold overlap to assess the degree of misalignment. The method can proceed up to step 324 in response to an overlap greater than the threshold overlap.

[0041] At 316, sample orientation is adjusted based on zone axis tilt. For example, the sample is tilted relative to two orthogonal axes by tilting the zone axis. After sample tilting, the sample height and / or the focus of the charged particle beam on the sample surface are adjusted.

[0042] Figure 5DThe diffraction pattern of the sample is shown after the sample orientation has been adjusted based on the zone axis tilt. The diffraction points of the Laue circle converge to and overlap with the direct beam. The centers of the Laue circle and the direct beam also overlap with each other.

[0043] At point 318, after tilting the sample, a second sample image is acquired, and the sample is offset based on a comparison between the second sample image and the initial sample image. This is achieved through operations such as... Figure 1 The sample holder 13 is used to offset the sample. A second sample image is acquired with the same parameters as the initial sample image acquired at 304. After inserting the HAADF detector into the beam path, the second image can be acquired via the HAADF detector. The sample is positioned based on the displacement between the second sample image and the initial sample image. Figure 4 Offset in the xy plane as shown. Offset the sample to compensate for sample drift caused by tilting the sample and mechanical offset caused by tilting. Step 318 can be skipped if the system has concentric holders in both orthogonal directions and sample drift is minimal. The concentric holders maintain the concentricity of the sample and do not cause sample drift / offset during sample tilting.

[0044] At 320, a third sample image can optionally be acquired, and the charged particle beam can be shifted based on a comparison between the third sample image and the initial sample image. This can be achieved through operation. Figure 1 The scanning coil 123 is used to achieve beam shifting. A third sample image is acquired using the same parameters as the initial sample image acquired at 304. The sample is then analyzed based on the displacement between the third sample image and the initial sample image. Figure 4 Offset in the xy-plane shown. Compared to offsetting the sample using a sample holder at 318, beam offset can achieve a higher resolution offset between the sample and the incident beam. This step is not necessary if the sample holder has sufficient offset accuracy.

[0045] If the zone axis tilt determined at 314 is less than the threshold zone axis tilt, a high-resolution image of the sample is acquired at 324. The resolution of the acquired sample image is higher than that of the sample image acquired by zone axis alignment or sample drift compensation (e.g., sample images acquired at 304, 318, and 320).

[0046] In some instances, the trained network can segment only the regions corresponding to the Laue circle. The center of the Laue circle can be determined based on the segmented regions. The center of the direct beam can be predetermined, such as the center of the diffraction pattern. The sample can be tilted based on the positions of the Laue circle center and the direct beam center in the diffraction pattern.

[0047] In this manner, the zone axis of a curved or bent sample is automatically aligned with the incident beam based on the convergent beam diffraction pattern of the sample. Since the convergent beam can have a nanometer-scale beam profile (or beam cross-section at the sample surface), crystal orientation can be aligned over a small, selected region. Using a trained network, the zone axis tilt is automatically determined based on the positions of the segmented straight beam and segmented Laue circles in the diffraction pattern. The quality of the network output is evaluated before tilting the sample to ensure the accuracy of the zone axis tilt estimation. Although zone axis alignment using a convergent beam diffraction pattern is provided as an example in this paper, method 300 can also achieve automatic zone axis alignment using a parallel beam diffraction pattern.

[0048] Figure 6 A method 600 for training a network with one or more reference samples is shown. The network may include a single machine learning network or multiple machine learning networks operating in combination. Individual machine learning networks may correspond to CAN, ANN, GAN, FCN, U-NET, YOLO, Mask R-CNN, or any other type of machine learning network capable of image segmentation. For example, the network may include a fully convolutional neural network. According to this disclosure, the network is trained with multiple diffraction patterns and annotated diffraction patterns of one or more reference samples.

[0049] The reference sample can be flat or have a surface smaller than 10 mm. Figure 3 The curvature of the sample being imaged. A flat reference sample ensures that the acquired diffraction pattern represents the specified tilt. Otherwise, due to sample curvature, if the sample shifts while the holder is tilted, the actual tilt may not match the specified tilt. If sample drift compensation is used, as described in step 608, the need for flatness is reduced. The sample material may or may not be the same as the material of the imaged sample. The reference and imaged samples are crystals of the same lattice type. For example, the reference and imaged samples may be silicon near the 110 zone axis. It doesn't matter whether the silicon has been processed by photolithography, etching, doping, or other semiconductor manufacturing processes. By including these lattice types during training, support for other lattice types can be achieved. The thickness of the reference sample may be different from or the same as the imaged sample if a beam of charged particles can be transmitted through the material.

[0050] At position 602, the zone axis of the reference sample is aligned with the incident beam axis. The zone axis of the reference sample can be manually aligned.

[0051] At position 604, acquire an initial image of the reference sample. The reference sample image can be a STEM image used to compensate for sample drift.

[0052] At 606, the reference sample is tilted within a predetermined tilt range for a predetermined step size, resulting in a diffraction pattern of the tilted reference sample. For example, the tilt step size can be 1 degree, and the tilt range in both the α-rotation direction and the β-rotation direction can be -5 degrees to 5 degrees. In another example, the tilt step size can be varied based on the total sample tilt angle. The tilt step size can be reduced at smaller tilt angles. This can be compared with... Figure 3 The same system configuration or parameters are used during zone axis alignment to obtain diffraction patterns. For example, with... Figure 3 The same beam convergence angle in 304 is used to obtain the diffraction pattern.

[0053] At position 608, a second reference sample image is acquired, and the reference sample is offset by comparing the second reference sample image with the initial reference sample image. Similar to... Figure 3 In step 314, at 606, the sample is offset in the xy plane to compensate for sample drift during sample tilting. Step 608 may further include offsetting the incident beam based on a comparison of the second reference sample image and the initial reference sample image.

[0054] At 610, method 600 checks whether a complete tilt sequence has been acquired. If a complete tilt sequence has been acquired, the method moves to 612. Otherwise, at 606, the sample is tilted further according to a predetermined step size.

[0055] At position 612, each diffraction pattern in the tilt sequence is annotated based on the known zone axis tilt corresponding to the diffraction pattern. Annotating the diffraction pattern involves masking the regions corresponding to the direct beam and the Laue circle within the diffraction pattern. The region corresponding to the direct beam can be annotated based on the known beam position in the diffraction pattern and the known beam size proportional to the beam convergence angle. The region corresponding to the Laue circle can be annotated based on the known zone axis tilt angle.

[0056] In one instance, such as Figure 7A The annotated diffraction pattern shows two circles representing the direct beam and the Laue circle. Circle 701 corresponds to the direct beam, and circle 702 corresponds to the Laue circle. The center position and radius of circle 702 are determined based on the known sample zone axis tilt from the incident beam and the beam convergence angle.

[0057] Return to Figure 6 At position 614, both the tilted sequence and the annotated tilted sequence are used to train the network. The network receives the diffraction pattern as input and outputs two segmented regions corresponding to the straight beam and the Laue circle. Figure 7B An exemplary output of the network is shown. The output is an image containing a first segmented region 704 corresponding to the Laue circle and a second segmented region 703 corresponding to the direct beam. This can be based on... Figure 7A Annotated diffraction patterns and Figure 7B The network parameters are adjusted based on the similarity of the network outputs. For example, similarity can be calculated based on the cross-entropy of two images. Training is not complete until the difference between the annotated diffraction pattern and the network output is less than a predetermined threshold level or when the difference no longer improves. The parameters of the trained network can be stored in the system's non-transitory memory for use in zone axis alignment. Figure 3 Method 300).

[0058] In some instances, the network is trained using multiple reference samples. These reference samples can have different thicknesses or different crystal orientations. They can be made of different materials or have different lattice types. In some instances, the network is trained on tilted sequences collected at different points on the reference samples.

[0059] In some instances, the Laue circle used for network training and stripline alignment is a zero-order Laue circle. In other instances, the Laue circle may contain higher-order Laue circles, such as a combination of a zero-order Laue circle and a first-order Laue circle.

[0060] In some instances, the network is trained to segment only the regions corresponding to the Laue circle, but not the regions corresponding to the direct beam.

[0061] In this way, a trained network for segmenting Laue circles and direct beams is generated based on multiple diffraction patterns from one or more reference samples. Annotations for each diffraction pattern are automatically generated based on the known zone axis tilt angle and the known beam convergence angle.

Claims

1. A method for zone axis alignment, comprising: Direct the charged particle beam toward the sample; Obtain the diffraction pattern of the sample; The trained network is applied to the diffraction pattern, where the output of the trained network corresponds to the first segmentation region of the Laue circle and the second segmentation region corresponding to the direct beam. as well as The zone axis tilt is determined based on the relative positions of the first segmented region and the second segmented region.

2. The method according to claim 1, further comprising: The sample is tilted based on the determined zone axis tilt.

3. The method according to claim 1, further comprising: The shape of the segmented region is determined, and in response to the shape of the segmented region being circular, the inclination of the zone axis is determined based on the segmented region, wherein the segmented region includes at least one of the first segmented region and the second segmented region.

4. The method according to claim 1, wherein, The charged particle beam is a converging charged particle beam.

5. The method according to claim 1, wherein, The charged particle beam is a parallel charged particle beam.

6. The method of claim 1, further comprising: The network is trained using multiple diffraction patterns obtained from the charged particle beam.

7. The method according to claim 1 or 6, further comprising: The network is trained using multiple diffraction patterns of a second sample, which has a smaller curvature than the first sample.

8. The method according to claim 7, wherein, Training the network using multiple diffraction patterns from the second sample includes: The plurality of diffraction patterns of the second sample are obtained by tilting the second sample along two orthogonal axes with a known tilt angle; Based on the known tilt angle, the direct beam and the Laue circle are annotated in the plurality of diffraction patterns; and The network is trained using multiple acquired diffraction patterns and multiple annotated diffraction patterns.

9. The method according to claim 8, wherein, Training the network using the acquired multiple diffraction patterns and the multiple annotated diffraction patterns includes: Input the plurality of diffraction patterns into the network; and The parameters of the network are adjusted by comparing the network's output with the plurality of annotated diffraction patterns.

10. The method according to any one of claims 1 to 3, further comprising: Determining a quality factor based on the first segmented region and the second segmented region; and determining the zone axis tilt based on the first segmented region and the second segmented region includes determining the zone axis tilt in response to the quality factor.

11. The method of claim 10, wherein determining the quality factor based on the first segmented region and the second segmented region includes determining the quality factor based on the degree of overlap between the first segmented region and the second segmented region.

12. The method of claim 10, wherein determining the quality factor based on the first segmented region and the second segmented region comprises determining the quality factor based on the shape of the first segmented region.

13. A system for imaging a sample, comprising: The source, which is used to generate beams of charged particles; A sample holder for tilting the sample; Detector; as well as A controller having instructions stored in a non-transitory memory, the controller being configured to: The charged particle beam is directed toward the sample; Obtain the diffraction pattern of the sample; The trained network is applied to the diffraction pattern, where the output of the trained network corresponds to the first segmentation region of the Laue circle and the second segmentation region corresponding to the direct beam. The zone axis tilt is determined based on the relative positions of the first segmented region and the second segmented region.

14. The system according to claim 13, wherein, The controller is further configured to: determine a quality factor based on the first segmented region; and tilt the sample in response to the quality factor.

15. The system according to any one of claims 13 to 14, wherein, The charged particle beam is a converging charged particle beam, and wherein the controller is further configured to tilt the sample based on a determined zone axis tilt; and to acquire an image of the sample by scanning the converging charged particle beam on the sample after tilting the sample.

Images