JP6225579B2 - Sample analysis program, sample analysis apparatus, and sample analysis method - Google Patents

Description

  The present invention relates to a sample analysis program, a sample analysis device, and a sample analysis method.

  One of the electron microscopes is a scanning transmission electron microscope (STEM). For example, a technique for analyzing a sample using a sample image (STEM image) obtained by such a STEM is known.

JP 2013-58363 A

  When analyzing a sample using the STEM, an image (calculated image) obtained by reproducing the STEM image is obtained from the STEM image acquired for the sample, and the sample is analyzed using the calculated image. May be effective.

  However, there are cases where a calculation image that accurately reproduces the STEM image cannot be obtained, and it takes a long time to obtain a calculation image that accurately reproduces the STEM image.

According to one aspect, the computer, using a scanning transmission electron microscope image of the sample, to produce an atomic placement information including coordinate information indicating the position of the atoms contained in the sample, wherein the atom placement generated Using the coordinate information included in a part of the range in the information, first information indicating a first calculation image obtained by calculating the position of the atom in the range is generated, and the first information is generated from the generated first information . Extracting the second information indicating the center of one calculation image, changing the range, generating the first information and extracting the second information, and using the extracted second information group A sample analysis program for executing a process for generating a second calculation image of a sample is provided.

Moreover , according to one viewpoint, the sample analysis apparatus and sample analysis method which produce | generate the calculation image of a sample are provided.

  According to the disclosed technology, it is possible to efficiently obtain a calculation image that accurately reproduces a STEM image. Further, by using such a calculation image, it is possible to improve the analysis accuracy of the sample.

It is a figure which shows the acquisition principle of the STEM image observed with a dark field scanning transmission electron microscope. It is a figure which shows an example of the hardware constitutions of the sample analysis system which concerns on embodiment. It is a figure which shows the structural example of STEM of the sample analysis system which concerns on embodiment. It is a figure which shows the scattering principle in the sample of the electron beam by STEM. It is a figure which shows an example of the functional block diagram showing the function with which the sample analyzer of the sample analysis system which concerns on embodiment is provided. It is a flowchart which shows an example of the sample analysis process performed with the sample analyzer which concerns on embodiment. It is a flowchart which shows an example of the calculation process of the STEM image performed with the sample analyzer which concerns on embodiment. It is a figure which shows an example of the STEM image of 1st Example. It is a figure for demonstrating the calculation process of the STEM image performed with the sample analyzer which concerns on 1st Example. It is a figure which shows the calculation image obtained with the sample analyzer which concerns on 1st Example. It is a figure which shows the calculation image with which the atom obtained by the sample analyzer which concerns on 2nd Example was missing. It is a figure which shows the calculation image to which the atom obtained with the sample analyzer which concerns on 2nd Example was added.

  First, the principle of acquiring a STEM image observed with an dark dark field (ADF) scanning transmission electron microscope (STEM) will be described.

FIG. 1 is a diagram showing the principle of acquiring a STEM image observed with a dark-field scanning transmission electron microscope.
The ADFSTEM 1000 includes at least an electron probe (not shown) that makes an electron beam incident on the sample 1100 and an annular detector 1400 that detects electrons 1300 diffracted (scattered) by the sample 1100. The detector 1400 is installed on a surface (a diffraction image surface) on which a diffraction image is formed from the detected electrons 1300. In addition, a sample analyzer (not shown) that performs control for observing the sample 1100 of the ADFSTEM 1000 and analyzes the sample 1100 is connected to the ADFSTEM 1000. The sample analyzer emits an electron beam from the electron probe, converges the diameter of the electron beam to nanometer or less (convergent electron beam 1200), and causes the converged electron beam 1200 to enter the sample 1100. In addition, the sample analyzer scans the main surface of the sample 1100 in the X direction and the Y direction (shown by dotted arrows in FIG. 2), and based on the electrons 1300 detected by the detector 1400. A diffraction image generation process is performed.

Using such an ADFSTEM 1000, the sample is analyzed as follows.
In the ADFSTEM 1000, when the sample 1100 is set at a predetermined observation position, the sample analyzer emits an electron beam from the electron probe, converges the electron beam to a desired diameter (convergent electron beam 1200), and converges electrons. A line 1200 is incident on the sample 1100. The focused electron beam 1200 is diffracted at an arbitrary angle by atoms contained in the sample 1100. The detector 1400 detects the number of electrons 1300 diffracted in a predetermined angle range among the electrons diffracted by the sample 1100 on the diffraction image plane. The sample analyzer synchronizes the number of electrons 1300 detected by the detector 1400 with the incident position of the convergent electron beam 1200 on the sample 1100, and uses the number as the image intensity. The sample analyzer scans the main surface of the sample 1100 with the focused electron beam 1200, and generates and acquires a two-dimensional dark field STEM image from the intensity of each image corresponding to each incident position of the focused electron beam 1200. .

  From this STEM image, the intensity corresponding to the atomic number of the atoms constituting the atomic column can be obtained at the position of the atomic column included in the sample 1100. Since the contrast of the STEM image hardly depends on the focal point of the electromagnetic lens that converges the electron beam from the electron probe onto the converged electron beam 1200 and the thickness of the sample 1100, the STEM image is excellent in atomic direct visibility. Further, the spatial resolution of the STEM image depends on the diameter for converging the convergent electron beam 1200. For example, when a spherical aberration corrector is used for converging the convergent electron beam 1200, the spatial resolution can be achieved to 1 nm or less. it can.

  The STEM image can be used, for example, for analysis of the defect structure of the sample 1100 and impurity atoms. In such analysis, a technique using an image acquired by theoretical analysis may be employed. For example, a calculation image obtained by reproducing the STEM image is obtained, and the calculation image is used for analysis of a defect structure or the like of the sample 1100.

  As a calculation method for acquiring a calculation image that reproduces it from an electron microscope image, there are a Bloch wave method and a multi-slice method. However, in these calculation methods, it may be difficult to obtain a highly accurate calculation image from the STEM image as described above. This point will be described below.

  The Bloch wave method uses the translational symmetry of crystals, Fourier transforms the crystal potential in real space to obtain the crystal potential in reciprocal lattice space, substitutes the obtained crystal potential into the Schrödinger equation, and Is calculated as a Bloch wave (wave function). Here, all diffracted waves generated in the reciprocal lattice space are taken into consideration, the eigenvalue calculation is performed on a square matrix of N rows and N columns corresponding to the number N of the waves, and a Bloch wave in the crystal is calculated. Using the Bloch waves thus obtained, the behavior of electrons on the lower surface of the crystal (the surface opposite to the incident surface of the electron beam) is obtained, the electrons passing through the crystal are calculated, and the diffraction image is calculated.

  In such a Bloch wave method, when the sample is a crystal having periodicity in a two-dimensional plane direction perpendicular to the incident direction of the electron beam, a periodic boundary condition can be set at the time of Fourier transform. By setting the periodic boundary condition, it is possible to obtain a calculation image without significantly increasing the number of waves to be considered.

  On the other hand, when the sample is a crystal having no periodicity in the two-dimensional plane direction perpendicular to the incident direction of the electron beam, it is difficult to set an appropriate periodic boundary condition at the time of Fourier transform. Setting a predetermined calculation range for calculation, for example, a superlattice range obtained by expanding the unit cell several times, and obtaining a calculation image of a dark field STEM image, discontinuity occurs at the boundary portion of the adjacent calculation range, An accurate calculation image may not be obtained. If the calculation is performed without setting the periodic boundary condition, all the reciprocal lattice points having the same resolution as that of the dark field STEM image must be considered. For example, when the calculation is performed on the dark field STEM image of 512 × 512 pixels, An eigenvalue calculation of a square matrix of 260,000 rows and 260,000 columns is required. This is the size of data that is practically impossible to calculate with current commercially available computers, and if a calculation is to be realized, a large-scale computing device, a huge memory, and a huge calculation time are required. For this reason, the Bloch wave method has hitherto been considered unsuitable as a calculation method for obtaining a calculation image used for analyzing a defect structure from a dark field STEM image.

  On the other hand, the multi-slice method divides a sample crystal into thin layers, calculates how the wave function propagates in the real space of each layer from the crystal top surface to the crystal bottom surface, and calculates the wave function on the crystal bottom surface. Ask for. In the multi-slice method, the propagation of electrons can be calculated with the crystal potential in the real space of each divided layer, and there is no need to consider the reciprocal space or to solve the Schrödinger equation. Without it, the wave function of the lower surface of the crystal can be calculated. The multi-slice method is one of calculation methods that can be used when acquiring a calculation image from the STEM image even when the sample is a crystal having no periodicity in a two-dimensional plane direction perpendicular to the incident direction of the electron beam. One.

  However, in the multi-slice method, when calculation is performed at one electron beam incident position and then at another electron beam incident position, it is necessary to calculate the propagation of the wave function from the crystal top surface to the crystal bottom surface again. is there. Therefore, when the resolution of the dark field STEM image corresponds to the incident position of the scanned electron beam as in the STEM in which the electron beam is scanned with respect to the sample, the calculation must be repeated by the number of resolutions. Calculation time becomes enormous. For example, even if the calculation time for one pixel of the dark field STEM image is 10 seconds, it takes about 30 days (= 512 × 512 × 10 [sec] / division) to calculate an image of 512 × 512 pixels. 86400 [sec / day]).

Therefore, in the following embodiment, a technique that can efficiently acquire a calculation image that accurately reproduces a STEM image will be described with reference to the drawings.
First, the sample analysis system will be described with reference to FIG.

FIG. 2 is a diagram illustrating an example of a hardware configuration of the sample analysis system according to the embodiment.
The sample analysis system 100 includes at least a STEM 200 for observing a sample, an input device 300, a display device 400, and a sample analysis device 500.

  The input device 300 is, for example, a keyboard or a mouse, and receives an operation input from a user and inputs an input signal to the sample analyzer 500. Specifically, the input device 300 causes the STEM 200 to emit an electron beam to the sample based on an operation input from the user, converge the emitted electron beam, and apply the converged electron beam to the sample. An input signal such as operation control related to observation of the STEM 200 such as scanning is input to the sample analyzer 500.

  The display device 400 is, for example, a display, and receives an image signal such as a STEM image observed by the STEM 200 and a calculation image acquired using the STEM image from the sample analysis device 500 and displays the image.

  The sample analysis apparatus 500 controls an electronic probe, a lens, a coil, or the like included in the STEM 200 based on an input signal from the input apparatus 300, causes the STEM 200 to perform sample observation, and displays an STEM image of the observed sample. To display. In addition, the sample analysis apparatus 500 performs a calculation process on the STEM image and generates a calculated image of the STEM image. Such a sample analyzer 500 includes a CPU (Central Processing Unit) 510a, a RAM (Random Access Memory) 510b, a HDD (Hard Disk Drive) 510c, a graphic processor 510d, and an input / output interface 510e. These parts are connected to each other by a bus 510f.

The CPU 510a controls the entire sample analyzer 500 by executing various programs stored in a storage medium such as the HDD 510c.
The RAM 510b temporarily stores at least a part of the OS and programs executed by the CPU 510a. The RAM 510b stores various data necessary for processing by the CPU 510a.

  The HDD 510c stores the OS and application programs on the sample analyzer 500. The HDD 510c stores various information necessary for processing by the CPU 510a.

  A display device 400 is connected to the graphic processing unit 510d. The graphic processing unit 510d causes the display device 400 to display a predetermined image in accordance with a command from the CPU 510a.

  The STEM 200 and the input device 300 are connected to the input / output interface 510e. The input / output interface 510e can be connected to a portable recording medium interface capable of writing information to the portable recording medium 300a and reading information from the portable recording medium 300a. The input / output interface 510e notifies the control signal from the CPU 510a to the STEM 200 via the bus 510f, and similarly notifies the CPU 510a of the signals from the STEM 200 and the input device 300.

Details of the functions of the sample analyzer 500 will be described later.
Next, details of the configuration of the STEM 200 included in the sample analysis system 100 will be described with reference to FIG.

FIG. 3 is a diagram illustrating a configuration example of the STEM of the sample analysis system according to the embodiment.
The STEM 200 includes an electron probe 201 that emits an electron beam B, a converging lens 202 that converges the electron beam B emitted from the electron probe 201, and a focal point of the electron beam B emitted from the electron probe 201 at the position of the sample S. And an objective lens 204 for focusing. An electromagnetic lens can be used for the converging lens 202 and the objective lens 204.

  The STEM 200 is arranged between the electron probe 201 and the objective lens 204 and scans the surface of the sample 210 with the electron beam B, and image formation that forms an image of the electron beam that has passed through the sample 210 on the image plane. And a lens 205.

  The STEM 200 includes a detector 206 that detects electrons scattered within a predetermined angle range from the optical axis OA. The detector 206 is a disc-shaped detector, and is disposed on an image forming plane on which a bright field image is formed.

  The STEM 200 includes a detector 207 that detects electrons diffracted (scattered) within a predetermined angle range from the optical axis OA. The detector 207 is an annular detector, and is disposed on an image forming plane on which a dark field image is formed.

In addition, the STEM 200 can include a converging lens diaphragm that cuts unnecessary spread of an electron beam or a correction unit that corrects aberrations such as spherical aberration and astigmatism.
Next, the principle of scattering of the electron beam B in the sample 210 by the STEM 200 having such a configuration will be described with reference to FIG.

FIG. 4 is a diagram showing the principle of scattering of an electron beam by a STEM.
FIG. 4 shows an example of the result of calculating the behavior of electrons in the crystal sample. FIG. 4 schematically shows a state where the incident electron beam B is scattered by the atomic column included in the sample 210 and the scattered electrons 213 are detected by the detector 207 in the STEM 200 as described above.

  In the STEM 200, when the electron beam B converged by the converging lens 202 (FIG. 3) is incident on the upper portion of the sample 210, the electrons 212 are localized in the atomic column 211 around the incident position of the electron beam B at a very high density. When the electron beam B converged using the spherical aberration correction apparatus is incident on the sample 210, the localization of the electrons 212 becomes more remarkable.

FIG. 4 shows an example of the result of calculation of the behavior of the electrons 212 in the sample 210, and the electrons 212 are hardly localized in atoms away from the incident position of the electron beam B.
The electrons 212 localized at the position of the atomic column 211 around the incident position of the electron beam B are inelastically scattered at a high angle by the thermal vibration of atoms constituting the atomic column 211. The electrons scattered by this electron scattering process are called thermal diffuse scattering (TDS) electrons 213.

  The TDS electrons 213 scattered at a high angle are detected by the detector 207. The scattering ability of the TDS electron 213 depends on the atomic number of the atom. For this reason, the STEM image generated by the sample analyzer 500 based on the TDS electrons 213 detected by the detector 207 shows the intensity depending on the atomic number.

Next, functions of the sample analysis apparatus 500 of the sample analysis system 100 will be described with reference to FIG.
FIG. 5 is a diagram illustrating an example of a functional block diagram illustrating functions included in the sample analysis device of the sample analysis system according to the embodiment.

  The sample analysis apparatus 500 of the sample analysis system 100 includes an image holding unit 501, an atomic information holding unit 502, a STEM control unit 503, an image generation unit 504, a calculation processing unit 505, and a display processing unit 506.

  The image holding unit 501 holds information indicating the STEM image observed by the STEM 200, information indicating the calculated image calculated by the calculation processing unit 505, and information indicating other images. Hereinafter, information indicating an image is also simply expressed as an image.

The atom information holding unit 502 holds atom arrangement information including coordinate information (XY coordinates) indicating the positions of atoms included in the sample 210.
The STEM control unit 503 controls each element illustrated in FIG. 3 so that the STEM 200 performs observation of the sample 210. For example, the STEM control unit 503 emits the electron beam B from the electron probe 201 of the STEM 200 and controls the converging lens 202 and the objective lens 204 to execute convergence of the diameter of the electron beam B and adjustment of the focus. In addition, the STEM control unit 503 controls the scanning coil 203 to cause the surface of the sample 210 to scan the electronic probe 201.

The image generation unit 504 generates a STEM image based on the number of electrons detected by the detector 207 by the STEM 200 and causes the image holding unit 501 to hold the STEM image.
The calculation processing unit 505 executes calculation processing for calculating a calculation image of the STEM image held in the image holding unit 501, and includes a peak search unit 505a, an image calculation unit 505b, and an image processing unit 505c. .

  The peak search unit 505a executes a peak search process on the STEM image of the sample 210, acquires atomic arrangement information of the sample 210 from the STEM image, and causes the atomic information holding unit 502 to hold it. Note that the peak search processing by the peak search unit 505a does not matter to the accuracy of the peak search, although there is a possibility that the acquisition of the atomic position may be missed or excessively acquired.

  The image calculation unit 505b designates a calculation range for a part of the atomic arrangement information acquired by the peak search unit 505a, performs calculation for the calculation range, and generates an image of the calculation range. The Bloch wave method can be used to calculate the image of the calculation range. When the image calculation unit 505b generates the image of the calculation range, the image calculation unit 505b moves the calculation range to another part in the atomic arrangement information, and the image of the calculation range is calculated for the calculation range of the movement destination in the same manner as described above. Generate. In this way, the image calculation unit 505b moves the calculation range within the atomic arrangement information, generates an image of each calculation range, and holds the image in the image holding unit 501.

In addition, the image calculation unit 505b extracts a trimmed image obtained by removing (trimming) other portions while leaving only a predetermined portion from the generated image in the calculation range.
As illustrated in FIG. 4, when the converged electron beam B is incident on the upper portion of the sample 210, the electron 212 is localized in the atomic column 211 around the incident position of the electron beam B, but is separated from the atomic column 211. Almost no locality. That is, it can be said that the atomic column 211 located away from the atomic column 211 in the vicinity of the incident position of the electron beam B does not contribute to the formation of the STEM image, or the influence thereof is extremely small.

  In addition, when a superlattice range is set as the calculation range, atoms different from the actual structure may appear at the end of the calculation range due to the non-periodicity of the superlattice (which may be a false image), but at the center Then that is unlikely to happen.

  The central portion of the calculation image in this calculation range corresponds to the vicinity of the electron beam B incident position at the time of STEM image acquisition, that is, the region where the electrons 212 are localized, and the influence from the peripheral region near the electron beam B incident position on the image formation. This corresponds to a region where there is no or is extremely small. The STEM image in the vicinity of the incident position of the electron beam B in which the influence from the peripheral region is suppressed coincides with the central portion of the calculation image in the calculation range with high accuracy. Therefore, among the calculation images in the calculation range, the region around the calculation image where atoms different from the actual structure may appear, and the region around the STEM image that hardly affects the imaging near the electron beam B incident position when acquiring the STEM image. The region around the calculated image corresponding to is trimmed. Then, the central portion of the calculated image corresponding to the vicinity of the incident position of the electron beam B at the time of acquiring the STEM image is left as a trimmed image. This makes it possible to generate a calculation image accurately reproduced from the STEM image.

  The size of the trimmed image can be set based on the diameter of the electron beam B incident on the sample 210 when the STEM image is acquired. For example, when spherical aberration correction is performed and the diameter of the electron beam B is about 0.2 nm, a rectangular region having a size with the diameter of the electron beam B as one side is extracted as a trimming image from the calculation image in the calculation range. To do. The size of the calculation range can be set to a range in which the peripheral region near the electron beam B incident position at the time of acquiring the STEM image does not affect the image formation near the incident position. The superlattice range of 0.8 nm can be set.

  The image processing unit 505c combines the trimmed images extracted from the images in the respective calculation ranges, combines the calculated images of the STEM images, and holds the combined images in the image holding unit 501. Further, the image processing unit 505c calculates the difference between the STEM image and the calculation image, and adds or deletes atoms from the calculation image according to the difference (coordinates indicating the atoms with respect to the information indicating the calculation image). Add or delete).

The display processing unit 506 displays the STEM image or calculation image held by the image holding unit 501 on the display device 400.
Note that the atomic information holding unit 502, the STEM control unit 503, the image generation unit 504, the calculation processing unit 505 (the peak search unit 505a, the image calculation unit 505b, the image processing unit 505c), and the display processing unit 506 provided in the sample analysis apparatus 500. The processing function is realized by executing the sample analysis program by the CPU 510a (FIG. 2).

Next, a sample analysis process executed by the sample analysis apparatus 500 will be described with reference to FIG.
FIG. 6 is a flowchart illustrating an example of a sample analysis process executed by the sample analysis apparatus according to the embodiment.

  When the user sets the sample 210 to be analyzed at a predetermined position in the STEM 200 and inputs an operation start of analysis to the sample analyzer 500, the sample analyzer 500 executes the following processing after initialization. .

  [Step S11] The STEM control unit 503 emits the electron beam B from the electron probe 201 of the STEM 200, controls the converging lens 202 and the objective lens 204, and converges the diameter of the electron beam B incident on the sample 210. The electron beam B is focused on the sample 210. Further, the STEM control unit 503 controls the magnetic field generated from the scanning coil 203 to scan the sample 210 with the electron beam B. The electron beam B incident and scanned in this manner is diffracted by the sample 210, and the diffracted electrons are detected by the detector 207.

  The image generation unit 504 acquires the number of electrons (image intensity) detected for each incident position of the electron beam B with respect to the sample 210 from the detector 207, and generates an STEM image. The image generation unit 504 causes the image holding unit 501 to hold the STEM image.

  [Step S12] The peak search unit 505a of the calculation processing unit 505 performs peak search processing on the STEM image held in the image holding unit 501, and acquires atomic arrangement information of the sample 210 from the STEM image.

The peak search unit 505a causes the atomic information holding unit 502 to hold this atomic arrangement information.
[Step S13] The calculation processing unit 505 executes calculation processing using the Bloch wave method for the calculation range specified in the atomic arrangement information held by the atomic information holding unit 502, generates an image of the calculation range, Extract a trimmed image from the image.

The calculation processing unit 505 causes the image holding unit 501 to hold the extracted trimmed image.
Details of the processing in step S13 will be described later.
[Step S14] The image processing unit 505c of the calculation processing unit 505 combines the trimming images for each calculation range held by the image holding unit 501, and synthesizes the calculation image of the STEM image.

The image processing unit 505c causes the image holding unit 501 to hold the combined calculation image.
[Step S15] The image processing unit 505c of the calculation processing unit 505 calculates a difference between the STEM image acquired in step S11 and the calculation image synthesized in step S14.

[Step S16] The image processing unit 505c of the calculation processing unit 505 calculates a difference in step S15 and determines whether or not the difference is obtained.
The image processing unit 505c executes the process of step S17 when the difference is obtained, and ends the sample analysis process when the difference is not obtained.

When the difference is obtained, the image processing unit 505c causes the image holding unit 501 to hold the difference.
[Step S17] The peak search unit 505a of the calculation processing unit 505 performs peak search processing on the difference between the STEM image held by the image holding unit 501 and the calculated image, and obtains position information indicating the position of the atom in the difference. Acquire differential atomic configuration information.

The peak search unit 505a causes the atomic information holding unit 502 to hold this difference atom arrangement information.
[Step S <b> 18] The image processing unit 505 c of the calculation processing unit 505 determines whether the calculation image synthesized in step S <b> 14 includes an atomic position acquisition omission or excessive acquisition. Therefore, it is determined whether all the atoms of the difference atom arrangement information are included in the STEM image or whether there are atoms not included in the difference atom arrangement information in the STEM image.

  The image processing unit 505c executes the process of step S19 when all the atoms of the difference atom arrangement information are included in the STEM image, that is, when acquisition of the position of the atom is included in the calculation image. On the other hand, if there is an atom that is not included in the difference atom arrangement information in the STEM image, that is, if the calculation image includes excessive acquisition of the position of the atom, the image processing unit 505c executes the process of step S20. To do.

  [Step S19] The image processing unit 505c of the calculation processing unit 505 adds an atom (coordinate indicating an atom) to the acquisition omission position in the calculation range (step S13) including the difference atom arrangement information acquired in step S17. Execute.

  [Step S20] The image processing unit 505c of the calculation processing unit 505, for the calculation range (step S13) including the difference atom arrangement information acquired in step S17, the atom (coordinates indicating the atom) from the position where atoms are acquired excessively. ) Is deleted.

  [Step S21] The image calculation unit 505b of the calculation processing unit 505 performs the calculation process again on the calculation range in which atoms are added in step S19 or the calculation range in which atoms are deleted in step S20, and generates an image of those calculation ranges. Then, a trimmed image is extracted from the image.

  The image calculation unit 505b holds the trimming image extracted from the recalculated calculation range image in the image holding unit 501, and the image calculation unit 501 holds the trimming image of the calculation range image before recalculation held by the image holding unit 501. Is updated.

In subsequent step S14, the image processing unit 505c resynthesizes the calculated image including the trimmed image in step S21.
Thus, the sample analysis process executed by the sample analysis apparatus 500 is completed.

Next, STEM image calculation processing (step S13) executed in the sample analysis processing of FIG. 6 will be described with reference to FIG.
FIG. 7 is a flowchart illustrating an example of STEM image calculation processing executed by the sample analyzer according to the embodiment.

[Step S13a] The image calculation unit 505b specifies a calculation range for the atomic arrangement information acquired from the STEM image in the peak search process.
[Step S13b] The image calculation unit 505b extracts atomic arrangement information of atoms within a specified calculation range.

[Step S13c] The image calculation unit 505b performs a calculation process on the calculation range based on the atomic arrangement information of the atoms extracted from the calculation range, and generates an image of the calculation range.
[Step S13d] The image calculation unit 505b trims the image in the calculation range and extracts a trimmed image.

[Step S13e] The image calculation unit 505b determines whether or not the calculation has been completed for all the atomic arrangement information acquired in the peak search process (step S12).
If all the calculations are not completed, the image calculation unit 505b executes the process of step S13f. If all the calculations are completed, the image calculation unit 505b ends the STEM image calculation process and executes the process of step S14.

[Step S13f] The image calculation unit 505b moves the calculation range with respect to the STEM image.
Thereafter, the image calculation unit 505b executes the process of step S13b again.

Thus, the STEM image calculation process in step S13 ends.
Note that the image calculation unit 505b can generate a plurality of calculation range images in accordance with steps S13a to 13c and 13f, and then extract a trimmed image from each of the calculation range images.

Next, the sample analysis process executed according to the flowcharts (FIGS. 6 and 7) will be specifically described.
(First embodiment)
In the first embodiment, a wide range of calculation images calculated by a system having a periodic structure in a two-dimensional plane perpendicular to the incident direction of the electron beam B are rotated by an appropriate angle so as to eliminate periodicity, and square The image obtained by cutting out into STEM images is an STEM image (original image).

  With respect to such a STEM image, a sample analysis process executed by the calculation processing unit 505 of the sample analyzer 500 will be described with reference to FIGS. 8 to 10 along the flowcharts of FIGS.

  FIG. 8 is a diagram illustrating an example of the STEM image of the first embodiment, and FIG. 9 is a diagram for explaining the calculation processing of the STEM image executed by the sample analyzer according to the first embodiment. FIG. 10 is a diagram showing a calculation image obtained by the sample analysis apparatus according to the first example. FIG. 10A shows a STEM image, FIG. 10B shows a calculation image of the STEM image, Each is shown.

First, the STEM image 10 of Si (silicon) (011) shown in FIG. 8 is held in advance in the image holding unit 501 of the sample analyzer 500.
The peak search unit 505a of the calculation processing unit 505 performs peak search processing on the STEM image 10 and acquires atomic arrangement information indicating the position of atoms from the STEM image 10 (FIG. 6: Step S12).

  In the first embodiment, since the calculation image is the STEM image 10 (original image), the atomic arrangement information indicating the positions of all atomic columns of the STEM image 10 is obtained by the peak search process without being affected by noise or the like. Can be acquired.

Next, the calculation processing unit 505 executes calculation processing of the STEM image 10 (FIG. 6: Step S13).
Specifically, first, as shown in FIG. 9A, the image calculation unit 505b of the calculation processing unit 505 has, for example, one side with respect to the atomic arrangement information 20 acquired from the STEM image 10 by the peak search process. The calculation range 21 which is a superlattice of about 0.8 nm is specified, and the atomic arrangement information of atoms in the calculation range 21 is extracted (FIG. 7: steps S13a and S13b).

  The image calculation unit 505b executes calculation processing for the calculation range 21 based on the atomic arrangement information of the atoms extracted from the calculation range 21, and generates an image 31 of the calculation range 21 as shown in FIG. 9B. (FIG. 7: Step S13c).

  The image calculation unit 505b trims a range other than the central portion where one side of the image 31 is about 0.2 nm and acquires a trimmed trimmed image 32a (see FIG. 9B) (FIG. 7: Step S13d). .

  The image calculation unit 505b determines whether or not the calculation has been completed for all the atomic arrangement information 20 acquired by the peak search process. If all the calculations have not been completed, the image calculation unit 505b determines from FIG. The calculation range 22 is newly designated by shifting 0.2 nm to the middle right (FIG. 7: Steps S13e and S13f).

  The image calculation unit 505b also acquires the trimming image 32a for the calculation range 22 from the image 32 generated by calculation processing in the same manner as in the calculation range 21 (FIG. 7: Steps S13b to S13d).

  The image calculation unit 505b determines whether or not the calculation is complete for the atomic arrangement information 20 acquired by the peak search process. If all the calculations are not completed, the image calculation unit 505b determines from FIG. The calculation range 23 is newly designated by shifting 0.2 nm to the middle right (FIG. 7: Steps S13e and S13f).

  The image calculation unit 505b also acquires the trimming image 33a from the image 33 generated by the calculation processing for the calculation range 23 as in the case of the calculation range 21 (FIG. 7: Steps S13b to S13d).

  The image calculation unit 505b moves the calculation range until the calculation of all the atomic arrangement information 20 is completed in this way, and acquires trimming images (steps S13e, S13f, S13b to S23d).

  Next, the image processing unit 505c puts together the trimmed images 31a to 33a,... For each calculation range of the atomic arrangement information 20 (FIG. 9E), and as shown in FIG. 30 are synthesized (FIG. 7: step S14).

Next, the image processing unit 505c calculates a difference between the STEM image 10 and the calculated image 30 corresponding to the STEM image 10 (FIG. 6: Steps S15 and S16).
As described above, in the first embodiment, the atomic arrangement information indicating the positions of all atomic columns of the STEM image 10 can be acquired by the peak search process without being affected by noise or the like. Therefore, it is determined that there is no difference between the STEM image 10 and the calculated image 30 with respect to the STEM image 10 (FIG. 6: step S16 (no)), and the sample analysis process ends.

  In the first embodiment, the sample analysis apparatus 500 acquires the atomic arrangement information 20 indicating the position of the atomic column from the STEM image 10, calculates the calculation range specified in the atomic arrangement information 20, and obtains an image of the calculation range. And trimming the generated image to extract a trimmed image. Further, the sample analysis apparatus 500 moves such a calculation range within the atomic arrangement information 20, extracts a trimming image from each calculation range, and combines the trimming images to generate a calculation image 30.

  At this time, the calculation range specified in the atom arrangement information 20 includes a false image in which the boundary portion includes atoms at positions where the atomic column does not originally exist, and the central portion includes the original atomic column. It is an image that accurately reflects the position of. Therefore, the trimmed image is obtained by trimming the calculation range and extracting the central portion. Since the calculated image 30 is a composite of such trimmed images, the STEM image 10 including the aperiodic structure is accurately reproduced.

  Further, the calculation image 30 is generated without performing a large-scale matrix operation or calculating a wave function from the upper surface to the lower surface of the sample for each electron beam incident position. Therefore, the amount of calculation required for generating the calculation image 30 is reduced, the calculation time is shortened, and the used capacity of the RAM 510b used for the calculation is also reduced.

(Second embodiment)
In the second embodiment, another specific example using the sample analysis process will be described with reference to FIGS. 11 and 12 according to the flowcharts of FIGS.

  11A and 11B are diagrams showing calculation images obtained by the sample analyzer according to the second embodiment in which atoms are missing. FIG. 11A shows a STEM image, and FIG. 11B shows a calculation image. A state in which the image is superimposed on the STEM image is shown.

  FIG. 12 is a diagram showing a calculation image in which atoms obtained by the sample analyzer according to the second embodiment are added. FIG. 12A shows an image of the difference between the STEM image and the calculation image. FIG. 12B shows a state in which the calculated image with the atoms added is superimposed on the STEM image.

When the user sets a sample at a predetermined position of the STEM 200 and inputs an operation start of analysis to the sample analysis apparatus 500, the sample analysis apparatus 500 executes observation of the sample.
The STEM control unit 503 causes the electron beam B to enter and scan the sample 210 in the STEM 200 to detect electrons diffracted by the sample 210. The image generation unit 504 acquires the number of electrons (image intensity) detected for each incident position of the electron beam B on the sample 210, and generates the STEM image 10a shown in FIG. Step S11).

  According to FIG. 11 (A), the STEM image 10a includes twin defects in a narrow range so as to straddle the central portion from the upper left to the lower right in FIG. 11 (A). Due to the twin defects, the periodicity in the two-dimensional plane direction perpendicular to the electron beam B is impaired.

  Next, the peak search unit 505a of the calculation processing unit 505 performs peak search processing on the STEM image 10a, and acquires atomic arrangement information indicating the position of atoms included in the sample 210 from the STEM image 10a (FIG. 6: step). S12).

  The calculation processing unit 505 executes calculation processing for a calculation range that is a range of a superlattice having a side of 0.8 nm specified in the acquired atomic arrangement information, generates an image of the calculation range, and trims the image A trimming image is extracted from a central portion having a side of 0.2 nm. The calculation processing unit 505 moves the calculation range until image calculation is completed for all the atomic arrangement information acquired by the peak search process, extracts a trimmed image in the same manner as described above, and holds the extracted image in the image holding unit 501 ( FIG. 6: Step S13, FIG. 7: Steps S13a to S13f).

The image processing unit 505c combines the trimmed images for each calculation range held by the image holding unit 501 and synthesizes the calculated image 30a of the STEM image (FIG. 6: step S14).
FIG. 11B shows a calculation image 30a superimposed on the STEM image 10a. According to this, it is recognized that the atoms in the part surrounded by the broken-line circle in FIG. This is because atomic arrangement information indicating the positions of some atoms included in the sample 210 could not be acquired from the STEM image 10a during the peak search process for the STEM image 10a.

  Next, the image processing unit 505c calculates a difference between the STEM image 10a and the calculated image 30a, executes a peak search process on the difference image, and obtains atomic arrangement information 40 indicating the position of the atom in the difference. Obtain (FIG. 6: Steps S15 to S17).

  FIG. 12A shows atomic arrangement information 40 indicating the positions of atoms in the difference between the STEM image 10a and the calculated image 30a. According to this, it can be seen that atoms exist at positions that could not be acquired by the peak search process for the STEM image 10a.

  The image processing unit 505c determines that the atomic arrangement information 40 is all included in the STEM image 10a, and adds atoms to the calculation range including the position of the atoms in the atomic arrangement information 40 (FIG. 6: Steps S18 and S19). .

  The image calculation unit 505b performs the calculation process again on the calculation range to which the atom at the position in the atom arrangement information 40 is added, generates an image of the calculation range, and extracts a trimmed image from the image. The image calculation unit 505b holds the trimming image extracted from the recalculated calculation range image in the image holding unit 501, and the image calculation unit 501 holds the trimming image of the calculation range image before recalculation held by the image holding unit 501. Is updated (FIG. 6: step S21).

The image processing unit 505c re-synthesizes the calculated image 30b using the updated trimmed image (FIG. 6: Step S14).
FIG. 12B shows a calculated image 30b superimposed on the STEM image 10a. According to this, it is recognized that the positions of the atoms in the calculation image 30b and the STEM image 10a coincide.

The image processing unit 505c determines that there is no difference between the STEM image 10a and the calculated image 30b (FIG. 6: step S16 (no)), and the sample analysis process ends.
Also in the second embodiment, similarly to the first embodiment, the sample analyzer 500 acquires the atomic arrangement information 20 indicating the position of the atomic column from the STEM image 10 and calculates the calculation range specified in the atomic arrangement information 20. Then, an image in the calculation range is generated, and the generated image is trimmed to extract a trimmed image. Furthermore, the sample analysis apparatus 500 can move such a calculation range within the atomic arrangement information 20, extract a trimming image from each calculation range, and combine the trimming images to generate the calculation image 30.

  At this time, the calculation range specified in the atomic arrangement information includes a false image in which the boundary part includes atoms at positions where the atomic column originally does not exist, and the central part includes the original atomic column. The image accurately reflects the position. Therefore, the trimmed image is obtained by trimming the calculation range and extracting the central portion.

  Further, when the atomic arrangement information indicating the position of the atomic column is acquired from the STEM image 10a, the difference between the STEM image 10a and the calculated image 30a of the STEM image 10a is obtained even if the acquisition of the position is omitted or excessive acquisition occurs. From the image, the atomic arrangement information 40 indicating the position of the acquisition missing or excessively acquired atomic column is acquired, and atoms are added to or deleted from the calculation range including the position indicated by the acquired atomic arrangement information 40.

  Therefore, since the calculation image 30b is a composite of trimming images extracted from the center portion of the image in such a calculation range, the STEM image 10a including the aperiodic structure is reproduced with higher accuracy. .

  Further, the calculation image 30b of the second embodiment is also generated without performing a large-scale matrix calculation or calculating a wave function from the upper surface to the lower surface of the sample for each incident position of the electron beam. For this reason, the calculation amount required for generating the calculation image 30b is reduced, the calculation time is shortened, and the used capacity of the RAM 510b used for the calculation is also reduced.

In the first and second embodiments described above, a Si stacking fault having a size of 8 nm on one side is calculated, and a very good match is found between the original image and a calculated image obtained using the same.
If calculation of this scale is performed using the conventional method using the Bloch wave method, it is necessary to perform eigenvalue calculation of 10,000 rows and 10,000 columns. This is a size that is practically impossible to calculate with current commercially available computers, and also requires a huge amount of memory. On the other hand, in the first and second embodiments described above, a superlattice having a size of 1/10 having a side of 0.8 nm is used for the calculation, and therefore, eigenvalue calculation of 1000 rows and 1000 columns is sufficient. This calculation scale can be said to be a very realistic calculation scale in analyzing a sample.

  Further, for example, when an image having a resolution of 896 pixels × 896 pixels is calculated by the multi-slice method, it is necessary to repeat the calculation for the number of pixels, and thus it is necessary to perform 802,816 repetitions. In the multi-slice method, even if the time required for calculation of one pixel is 10 seconds, calculation of an image with the above resolution requires calculation time of 80 days or more. On the other hand, in the first and second embodiments, the calculation required to obtain the calculation image can be completed in about 6 days, and the efficiency can be increased by 10 times or more.

  The above processing functions can be realized by a computer. In that case, a program describing the processing contents of the functions that the sample analyzer 500 should have is provided. By executing the program on a computer, the above processing functions are realized on the computer. The program describing the processing contents can be recorded on a computer-readable recording medium (including a portable recording medium). Examples of the computer-readable recording medium include a magnetic recording device, an optical disk, a magneto-optical recording medium, and a semiconductor memory. Examples of the magnetic recording device include a hard disk drive (HDD), a flexible disk (FD), and a magnetic tape. Optical discs include DVD (Digital Versatile Disc), DVD-RAM, CD-ROM (Read Only Memory), CD-R (Recordable) / RW (ReWritable), and the like. Magneto-optical recording media include MO (Magneto-Optical disk).

  When distributing the program, for example, a portable recording medium such as a DVD or a CD-ROM in which the program is recorded is sold. It is also possible to store the program in a storage device of a server computer and transfer the program from the server computer to another computer via a network.

  The computer that executes the program stores, for example, the program recorded on the portable recording medium or the program transferred from the server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read the program directly from the portable recording medium and execute processing according to the program. Further, each time the program is transferred from the server computer, the computer can sequentially execute processing according to the received program.

Regarding the embodiment described above, the following additional notes are further disclosed.
(Supplementary note 1)
Using the scanning transmission electron microscope image of the sample, generate atomic arrangement information including coordinate information indicating the position of the atoms contained in the sample,
Using the coordinate information included in a partial range in the generated atomic arrangement information, to generate first information indicating an image of the range,
From the generated first information, second information indicating a central portion of the image of the range is extracted,
Changing the range, generating the first information and extracting the second information;
A sample analysis program for executing a process of generating a calculated image of the sample using the extracted second information group.

(Additional remark 2) The sample analysis program of Additional remark 1 characterized by producing | generating the said calculation image by sticking the image which each of said 2nd information group shows.
(Additional remark 3) The center part of the image of the said range is set based on the diameter of the electron beam irradiated to the said sample at the time of the said scanning transmission electron microscope image acquisition, The additional remark 1 or 2 characterized by the above-mentioned. Sample analysis program.

(Additional remark 4) The said range is prescribed | regulated by the superlattice of the said sample, The sample analysis program in any one of Additional remark 1 thru | or 3 characterized by the above-mentioned.
(Supplementary note 5) The sample analysis program according to any one of supplementary notes 1 to 4, wherein the first information is generated using a Bloch wave method using the coordinate information included in the range.

(Appendix 6) In the computer,
Generating third information indicating a difference between the scanning transmission electron microscope image and the calculated image;
The sample analysis program according to any one of appendices 1 to 5, further comprising executing a process of correcting the calculation image using the third information.

(Supplementary Note 7) The process of correcting the calculation image is as follows.
Using the third information, for the difference range in which the difference exists in the atomic arrangement information, the coordinate information included in the difference range is used to generate fourth information indicating an image of the difference range,
From the fourth information, extract fifth information indicating a central portion of the image of the difference range,
The sample analysis program according to appendix 6, including a process of correcting the calculation image using the fifth information.

(Additional remark 8) The 1st production | generation part which produces | generates the atomic arrangement | positioning information containing the coordinate information which shows the position of the atom contained in the said sample using the scanning transmission electron microscope image of a sample,
A second generation unit that generates first information indicating an image of the range using the coordinate information included in a partial range in the atomic arrangement information generated by the first generation unit;
An extraction unit that extracts second information indicating a central portion of the image of the range from the first information generated by the second generation unit;
And a third generation unit that generates a calculated image of the sample using the second information extracted by the extraction unit.

(Additional remark 9) The process part which changes the said range is further included,
The second generation unit generates the first information indicating the image of the changed range using the coordinate information included in the range changed by the processing unit,
The extraction unit extracts the second information indicating the central part of the image of the changed range from the first information generated by the second generation unit,
The sample analysis apparatus according to appendix 8, wherein the third generation unit generates the calculation image using the second information group extracted by the extraction unit.

(Appendix 10) The computer
Using the scanning transmission electron microscope image of the sample, generate atomic arrangement information including coordinate information indicating the position of the atoms contained in the sample,
Using the coordinate information included in a partial range in the generated atomic arrangement information, to generate first information indicating an image of the range,
From the generated first information, second information indicating a central portion of the image of the range is extracted,
Changing the range, generating the first information and extracting the second information;
A sample analysis method, wherein a calculated image of the sample is generated using the extracted second information group.

100 Sample analysis system 200 STEM
DESCRIPTION OF SYMBOLS 201 Electronic probe 202 Converging lens 203 Scanning coil 204 Objective lens 205 Imaging lens 206,207,1400 Detector 210,1100 Sample 211 Atomic column 212,213,1300 Electronic 300 Input device 300a Portable recording medium 400 Display device 500 Sample analysis Apparatus 501 Image holding unit 502 Atomic information holding unit 503 STEM control unit 504 Image generation unit 505 Calculation processing unit 505a Peak search unit 505b Image calculation unit 505c Image processing unit 506 Display processing unit 510a CPU
510b RAM
510c HDD
510d Graphic processing unit 510e I / O interface 510f Bus 1000 ADFSTEM
1200 Convergent electron beam

Claims (6)

On the computer,
Using the scanning transmission electron microscope image of the sample, generate atomic arrangement information including coordinate information indicating the position of the atoms contained in the sample,
Using the coordinate information included in a partial range in the generated atomic arrangement information, to generate first information indicating a first calculation image that calculates the position of the atoms in the range,
From the generated first information, second information indicating a central portion of the first calculation image is extracted,
Changing the range, generating the first information and extracting the second information;
A sample analysis program for executing a process of generating a second calculated image of the sample using the extracted second information group. 2. The sample analysis program according to claim 1, wherein the second calculation image is generated by pasting a central portion of the first calculation image indicated by each of the second information groups. 3. The sample according to claim 1, wherein a central portion of the first calculation image is set based on a diameter of an electron beam irradiated on the sample when the scanning transmission electron microscope image is acquired. Analysis program. In the computer,
Generating third information indicating a difference between the scanning transmission electron microscope image and the second calculation image;
The sample analysis program according to any one of claims 1 to 3, further comprising executing a process of correcting the second calculation image using the third information. A first generation unit that generates atomic arrangement information including coordinate information indicating a position of an atom included in the sample, using a scanning transmission electron microscope image of the sample;
Using the coordinate information included in a partial range in the atomic arrangement information generated by the first generation unit, first information indicating a first calculation image obtained by calculating the position of the atom in the range is generated. A second generation unit;
An extraction unit for extracting second information indicating a central portion of the first calculation image from the first information generated by the second generation unit;
And a third generation unit that generates a second calculated image of the sample using the second information extracted by the extraction unit. Computer
Using the scanning transmission electron microscope image of the sample, generate atomic arrangement information including coordinate information indicating the position of the atoms contained in the sample,
Using the coordinate information included in a partial range in the generated atomic arrangement information, to generate first information indicating a first calculation image that calculates the position of the atoms in the range,
From the generated first information, second information indicating a central portion of the first calculation image is extracted,
Changing the range, generating the first information and extracting the second information;
A sample analysis method, wherein a second calculated image of the sample is generated using the extracted second information group.

Images