Electron microscope and its control method

Abstract

This invention provides an electron microscope and a control method thereof that can remove capsules from observed images without reproducing phase information. [Solution] An electron microscope comprising an electron source that emits an electron beam to irradiate a sample, a detector that detects electrons emitted from the sample and surrounding objects arranged around the sample, and a control unit that acquires an observation image based on the detection signal output from the detector, wherein the control unit acquires the observation image for each direction of the electron beam by controlling the direction of the electron beam relative to the sample, and removes the image of the surrounding objects from the observation image using an averaged image obtained by averaging the observation images.

Description

Technical Field

[0001] The present invention relates to an electron microscope used for observing a sample disposed in a capsule in which gas or liquid is enclosed, and a control method thereof.

Background Art

[0002] An electron microscope is a device that irradiates a sample disposed in a vacuum with an electron beam, detects secondary electrons, reflected electrons, and transmitted electrons generated thereby, and generates an observation image of the sample based on the detection signal. When observing a sample in gas or liquid, the sample is disposed in a capsule in which gas or liquid is enclosed. However, when a capsule is used, the electron beam irradiates not only the sample but also the capsule, so the observation image includes not only the sample but also the capsule. Since the capsule included in the observation image hinders the observation of the sample, it is preferably removed.

[0003] Patent Document 1 discloses reproducing amplitude information and phase information from a plurality of hologram images obtained by changing the incident angle of an electron beam, and separating and removing a capsule by an analysis algorithm of electron beam tomography.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in Patent Document 1, in order to separate and remove a capsule, it is essential to reproduce phase information from a hologram image or the like. In order to reproduce phase information, it is necessary to provide an electron beam biprism or the like, but since a general electron microscope does not have an electron beam biprism or the like, it is difficult to reproduce phase information.

[0006] Therefore, the present invention aims to provide an electron microscope and a control method thereof that can remove capsules from an observed image without reproducing phase information. [Means for solving the problem]

[0007] To achieve the above objective, the present invention provides an electron microscope comprising: an electron source that emits an electron beam to irradiate a sample; a detector that detects electrons emitted from the sample and surrounding objects arranged around the sample; and a control unit that acquires an observation image based on a detection signal output from the detector, wherein the control unit acquires an observation image for each direction of the electron beam by controlling the direction of the electron beam relative to the sample, and removes images of surrounding objects from the observation image using an averaged image obtained by averaging the observation images.

[0008] The present invention also relates to a control method for an electron microscope comprising an electron source that emits an electron beam to irradiate a sample, a detector that detects electrons emitted from the sample and objects surrounding the sample arranged around the sample, and a control unit that acquires an observation image based on a detection signal output from the detector and controls the operation of each part, wherein the control unit acquires an observation image for each direction of the electron beam by controlling the direction of the electron beam relative to the sample, and removes images of objects surrounding the sample from the observation image using an averaged image obtained by averaging the observation images. [Effects of the Invention]

[0009] According to the present invention, it is possible to provide an electron microscope and a control method thereof that can remove capsules from an observed image without reproducing phase information. [Brief explanation of the drawing]

[0010] [Figure 1] Overall configuration diagram of the electron microscope in Example 1 [Figure 2] A diagram showing an example of an observation image of a sample placed inside a capsule. [Figure 3] A diagram showing an example of the processing flow in Example 1. [Figure 4] Diagram illustrating the direction of the electron beam relative to the sample. [Figure 5] A diagram showing that the direction of the electron beam relative to the sample is set by the sample stage. [Figure 6] Figure 3 provides supplementary explanations. [Figure 7] A diagram showing another example of the processing flow in Example 1. [Figure 8] Figure 7 provides supplementary explanations. [Figure 9] Overall configuration diagram of the electron microscope in Example 2 [Modes for carrying out the invention]

[0011] The following describes embodiments of the electron microscope according to the present invention with reference to the attached drawings. An electron microscope is a device that detects secondary electrons, backscattered electrons, and transmitted electrons generated from a sample by irradiating the sample with an electron beam, and generates an observation image based on the detected signals. In addition, objects other than the sample may be placed around the sample. That is, objects around the sample are not the object of observation, but for example, capsules, films, sheets, meshes, transmission windows, membranes, etc. [Examples]

[0012] The overall configuration of the scanning electron microscope, which is the electron microscope of Example 1, will be explained using Figure 1. The scanning electron microscope comprises a microscope body 101 and a control unit 110.

[0013] The microscope body 101 includes an electron source 102, a deflector 104, a focusing lens 105, a sample stage 108, and a detector 109. The inside of the microscope body 101 is evacuated by a vacuum pump or the like, and a capsule 106 containing a gas or liquid is held on the sample stage 108 along with the sample 107. The capsule 106 is an example of an object surrounding the sample.

[0014] The electron source 102 emits an electron beam 103 that irradiates the capsule 106 and the sample 107. The deflector 104 deflects the electron beam 103 so as to scan the observation region. The condenser lens 105 converges the deflected electron beam 103 onto the observation region. The sample stage 108 that holds the capsule 106 containing the sample 107 moves horizontally and vertically to set the observation region at a desired position. Also, the sample stage 108 tilts as necessary. The detector 109 detects secondary electrons and reflected electrons emitted from the capsule 106 and the sample 107 due to the irradiation of the electron beam 103, and transmits a detection signal to the control unit 110.

[0015] The control unit 110 is, for example, a computer, generates an observation image based on the detection signal transmitted from the detector 109, and controls the operations of each part provided in the microscope body 101. A storage unit 111 and a display unit 112 are connected to the control unit 110. The storage unit 111 is, for example, an HDD (Hard Disk Drive) or an SSD (Solid State Drive), and stores various data related to the generation of the observation image and the control of the operations of each part, the observation image, and various images generated using the observation image. The display unit 112 is, for example, a liquid crystal display, and displays the observation image and various images.

[0016] An example of an observation image of the sample 107 disposed in the capsule 106 will be described using FIG. 2. The capsule 106 is composed of a carbon film reinforced with a microgrid, and the sample 107 is a latex sphere. The observation image 200 illustrated in FIG. 2 includes a black circular sample 107 located at the center and a mesh-like microgrid 201. Note that the carbon film, which is a light element and has a thin film thickness, is not included in the observation image 200.

[0017] The microgrid 201 included in the observation image 200 becomes noise when observing the sample 107 and hinders the observation of the sample 107. Therefore, in Example 1, by controlling the direction of the electron beam 103 with respect to the sample 107, an observation image is acquired for each direction of the electron beam 103, and a sample image from which the capsule 106 has been removed is generated using an averaged image obtained by averaging the plurality of acquired observation images.

[0018] Using FIG. 3, an example of the processing flow of Example 1 will be described for each processing step.

[0019] (S301) The control unit 110 sets the direction of the electron beam 103 with respect to the sample 107 by controlling the deflector 104. Specifically, the direction of the electron beam 103 with respect to the sample 107 is set by the control unit 110 controlling the current or voltage supplied to the deflector 104, which is a coil or electrode pair.

[0020] The direction of the electron beam 103 with respect to the sample 107 will be described using FIG. 4. The sample surface 400 including the observation region is a plane including the X-axis and the Y-axis, and the axis orthogonal to the sample surface 400 is the Z-axis. The direction of the electron beam 103 with respect to the sample 107 is determined by the incident angle θ and the azimuth angle φ. The incident angle θ is the angle formed by the Z-axis and the electron beam 103, and the azimuth angle φ is the angle formed by the electron beam 103 projected onto the XY plane and the X-axis. The electron beam 103 projected onto the XY plane is indicated by a dotted line in FIG. 4.

[0021] Also, as illustrated in FIG. 5, the direction of the electron beam 103 with respect to the sample 107 may be set by tilting the sample stage 108. The tilt angle of the sample stage 108 is controlled by the control unit 110. When the sample stage 108 is used for setting the direction of the electron beam 103, the direction of the electron beam 103 can be set in a wider range, and the influence of the aberration of the condenser lens 105 can be reduced. When the deflector 104 is used for setting the direction of the electron beam 103, the direction of the electron beam 103 can be set more finely.

[0022] (S302) The control unit 110 acquires an observation image in the direction of the electron beam 103 with respect to the sample 107 set in S301. The acquired observation image is stored in the storage unit 111.

[0023] (S303) The control unit 110 determines whether the termination condition has been met. If the termination condition is met, the process proceeds to S304; otherwise, the process returns to S301. In other words, the setting of the electron beam 103 direction in S301 and the acquisition of the observation image in S302 are repeated until the termination condition is met.

[0024] The termination conditions include, for example, the number of observation images acquired for each direction of the electron beam 103 exceeding a predetermined number, or the time required for setting the direction of the electron beam 103 and acquiring the observation images exceeding a predetermined time. Alternatively, the termination condition may be when the operator issues a command to terminate the repetition of steps S301 and S302.

[0025] Figure 6 shows examples of multiple observation images acquired in S301 to S303. Since the incident angle θ and azimuth angle φ are different for each observation image, the position of the mesh-like microgrid relative to the black circular sample is different. Alternatively, multiple observation images may be acquired by keeping the azimuth angle φ constant and only changing the incident angle θ. When only the incident angle θ is changed, the setting of the direction of the electron beam 103 relative to the sample 107 can be simplified.

[0026] (S304) The control unit 110 generates an averaged image by averaging the multiple observation images acquired in S301 to S303. Prior to generating the averaged image, each of the multiple observation images may be subjected to a positioning process based on a portion of the sample 107. For example, template matching may be used for the positioning process.

[0027] Figure 6 illustrates the averaged image generated in S304. Microgrids that are positioned differently relative to sample 107 in each observation image are dispersed and become indistinct in the averaged image. On the other hand, sample 107 is in the same position in each observation image and therefore becomes clear in the averaged image. Furthermore, sample 107 becomes even clearer in the averaged image generated after alignment processing for each observation image.

[0028] (S305) The control unit 110 generates multiple capsule images by subtracting the averaged image generated in S304 from each of the multiple observation images acquired in S301 to S303. That is, by subtracting the averaged image in which the sample 107 is clear and the microgrid is unclear from each observation image, a capsule image is generated in which the sample 107 is removed and the capsule 106 is imaged.

[0029] Figure 6 illustrates several capsule images generated in S305. Each capsule image includes only the microgrid, which is part of capsule 106, and does not include the sample 107. Since the position of the microgrid differs in each observation image, the position of the microgrid also differs in each capsule image. Furthermore, each capsule image includes the microgrid of the upper capsule, which is capsule 106 located closer to the electron source 102 than the sample 107, and the microgrid of the lower capsule, which is capsule 106 located closer to the sample stage 108 than the sample 107.

[0030] (S306) The control unit 110 generates an upper capsule image and a lower capsule image by performing a positional adjustment on the multiple capsule images generated in S305 and then averaging them. The upper capsule image is generated by performing a positional adjustment on the multiple capsule images based on the microgrid of the upper capsule and then averaging them. The lower capsule image is generated by performing a positional adjustment on the multiple capsule images based on the microgrid of the lower capsule and then averaging them. Template matching, for example, is used for the positional adjustment.

[0031] Figure 6 illustrates the upper and lower capsule images generated in S306. The upper capsule image includes the microgrid of the upper capsule located closer to the electron source 102 than to the sample 107, while the lower capsule image includes the microgrid of the lower capsule located closer to the sample stage 108 than to the sample 107.

[0032] (S307) The control unit 110 generates multiple sample images by subtracting the upper capsule image and the lower capsule image generated in S306 from each of the multiple observation images acquired in S301 to S303. That is, by subtracting the upper capsule image, which includes the microgrid of the upper capsule, and the lower capsule image, which includes the microgrid of the lower capsule, from each observation image, a sample image from which capsule 106 has been removed is generated. Note that generating multiple sample images is not mandatory; a single sample image may be generated by subtracting the upper capsule image and the lower capsule image from any of the multiple observation images.

[0033] Figure 6 illustrates several sample images generated in S307. In each sample image, the microgrids of the upper and lower capsules are removed, and sample 107 is imaged more clearly.

[0034] (S308) The control unit 110 reduces the noise contained in each sample image by averaging the multiple sample images generated in S307. Note that S308 is not mandatory, and if a single sample image is generated in S307, S308 is skipped.

[0035] As explained using Figure 3, the process generates a sample image from which the microgrid, which is part of the capsule 106, has been removed from the observation image which includes the sample 107 along with the microgrid. By removing the microgrid, the sample 107 can be observed in detail.

[0036] Using Figure 7, another example of the processing flow of Example 1 will be explained. Note that S301 to S307 are the same as in Figure 3, so S701 to S704, which are replaced by S308, will be explained below.

[0037] (S701) The control unit 110 determines whether or not there is a machine learning model that has learned noise reduction. If there is a machine learning model, processing proceeds to S702; otherwise, processing proceeds to S702 via S703.

[0038] (S702) The control unit 110 generates multiple denoised sample images by performing noise reduction using a machine learning model on each of the multiple sample images generated in S307. Note that generating multiple denoised sample images is not mandatory; a single denoised sample image may be generated by performing noise reduction using a machine learning model on any of the multiple observation images.

[0039] Figure 8 shows examples of multiple sample images generated in S307, along with multiple denoised sample images generated in S702. In the denoised sample image, where granular noise contained in the sample image has been removed, sample 107 is imaged more clearly.

[0040] (S703) The control unit 110 generates a machine learning model to perform noise reduction. Machine learning models such as DuCNN, Deep Image Prior, Noise2Noise, and Noise2Void are used. Multiple sample images generated in S307 are used as input and training images for generating the machine learning model. The generated machine learning model is stored in the memory unit 111.

[0041] (S704) The control unit 110 further reduces the noise remaining in each denoised sample image by averaging the multiple denoised sample images generated in S702. Note that S704 is not mandatory, and if a single denoised sample image is generated in S702, S704 is skipped.

[0042] As explained using Figure 8, a sample image is generated from the observation image which includes the sample 107 along with the microgrid, which is part of the capsule 106, similar to Figure 3. The removal of the microgrid allows for detailed observation of the sample 107.

[0043] Furthermore, since noise reduction using a machine learning model is performed on the sample image from which the microgrid has been removed, it is possible to generate a denoised sample image in which sample 107 is more clearly imaged. In addition, noise can be further reduced by averaging multiple denoised sample images. [Examples]

[0044] Example 1 described how to remove capsule-derived signals from observation images acquired by a scanning electron microscope. Example 2 describes how to remove capsule-derived signals from observation images acquired by a transmission electron microscope.

[0045] The overall configuration of the transmission electron microscope, which is the electron microscope of Example 2, will be explained using Figure 9. The transmission electron microscope comprises a microscope body 901 and a control unit 110.

[0046] The microscope body 901 includes an electron source 902, a deflector 904, a condenser lens 905, a sample stage 908, an objective lens 910, and a detector 909. The inside of the microscope body 901 is evacuated by a vacuum pump or the like, and the sample stage 908 holds a capsule 106 containing a gas or liquid along with the sample 107.

[0047] The electron source 902 emits an electron beam 903 that is irradiated onto the capsule 106 and the sample 107. The deflector 904 deflects the electron beam 903 to set the direction of the electron beam 903 relative to the sample 107. The condenser lens 905 shapes the electron beam 903. The sample stage 908 holds the capsule 106 containing the sample 107 and moves horizontally and vertically to set the observation area to a desired position. The sample stage 908 also tilts to set the direction of the electron beam 903 relative to the sample 107. The objective lens 910 magnifies the transmitted electrons, which are electrons that have passed through the capsule 106 and the sample 107, and images the transmitted electrons with the detector 909. The detector 909 detects the transmitted electrons and transmits a detection signal to the control unit 110.

[0048] The control unit 110, as in Embodiment 1, is, for example, a computer, and generates an observation image of the sample 107 based on the detection signal transmitted from the detector 909, and controls the operation of each part provided on the microscope body 901.

[0049] The processing flow in Example 2 is the same as in Example 1 and is illustrated in Figure 3 or Figure 7. Specifically, the control unit 110 controls the deflector 904 and the sample stage 908 to set the direction of the electron beam 903 relative to the sample 107, and a sample image with the capsule 106 removed is generated using an averaged image of the observation images acquired for each direction of the electron beam 903. Furthermore, a denoised sample image is generated by noise reduction using a machine learning model. In addition, noise is further reduced by averaging multiple denoised sample images.

[0050] Multiple embodiments of the electron microscope of the present invention have been described above. The present invention is not limited to the above embodiments, and the components can be modified and implemented without departing from the spirit of the invention. Furthermore, the multiple components disclosed in the above embodiments may be combined as appropriate. In addition, some components may be deleted from all the components shown in the above embodiments. [Explanation of Symbols]

[0051] Microscope body 101, electron source 102, electron beam 103, deflector 104, focusing lens 105, capsule 106, sample 107, sample stage 108, detector 109, control unit 110, memory unit 111, display unit 112, observation image 200, microgrid 201, sample surface 400, microscope body 901, electron source 902, electron beam 903, deflector 904, condenser lens 905, sample stage 908, detector 909, objective lens 910.

Claims

1. An electron microscope comprising an electron source that emits an electron beam to irradiate a sample, a detector that detects electrons emitted from the sample and surrounding objects arranged around the sample, and a control unit that acquires an observation image based on the detection signal output from the detector, The electron microscope is characterized in that the control unit controls the direction of the electron beam relative to the sample, thereby acquiring the observation image for each direction of the electron beam, and using the averaged image obtained by averaging the observation images to remove images of objects surrounding the sample from the observation image.

2. An electron microscope according to claim 1, The electron microscope is characterized in that the control unit controls the direction of the electron beam relative to the sample using a deflector that deflects the electron beam.

3. An electron microscope according to claim 1, The electron microscope is characterized in that the control unit controls the direction of the electron beam relative to the sample by tilting the sample stage that holds the sample.

4. An electron microscope according to claim 1, An electron microscope characterized in that the direction of the electron beam relative to the sample is determined by the incident angle and azimuth angle of the electron beam.

5. An electron microscope according to claim 4, The control unit is characterized by maintaining the azimuth angle constant and changing the incident angle.

6. An electron microscope according to claim 1, The electron microscope is characterized in that the control unit generates multiple images of objects surrounding the sample by subtracting the averaged image from each of the observation images acquired for each direction of the electron beam, generates an upper image of objects surrounding the sample and a lower image of objects surrounding the sample by performing a positional adjustment on the multiple images of objects surrounding the sample and then averaging them, and generates a sample image by subtracting the upper image of objects surrounding the sample and the lower image of objects surrounding the sample from the observation image.

7. An electron microscope according to claim 6, The electron microscope is characterized in that the control unit generates a plurality of sample images by subtracting the upper sample peripheral image and the lower sample peripheral image from each of the observed images, and reduces noise by averaging the plurality of sample images.

8. An electron microscope according to claim 6, The control unit is characterized by removing noise from the sample image using a machine learning model for noise removal.

9. An electron microscope according to claim 8, The electron microscope is characterized in that the control unit generates a plurality of sample images by subtracting the upper sample peripheral image and the lower sample peripheral image from each of the observed images, and uses the plurality of sample images to generate the machine learning model.

10. An electron microscope according to claim 1, The electron microscope is characterized in that the sample surrounding material is a capsule for enclosing the sample, a membrane covering the sample, or a sample covering for the sample.

11. A control method for an electron microscope comprising an electron source that emits an electron beam to irradiate a sample, a detector that detects electrons emitted from the sample and surrounding objects arranged around the sample, and a control unit that acquires an observation image based on the detection signal output from the detector and controls the operation of each part, The control unit controls the direction of the electron beam relative to the sample, thereby acquiring the observation image for each direction of the electron beam, and using the averaged image obtained by averaging the observation images, removes images of objects surrounding the sample from the observation image.

Images