JP7701601B2 - Method for observing magnetic domains in steel sheets - Google Patents

Description

The present invention relates to a method for observing magnetic domains in steel sheets.

It is required to achieve low iron loss characteristics and to control noise problems caused by magnetostriction in electrical steel sheets. Iron loss is the power loss consumed as heat energy when an iron core is excited with an alternating magnetic field, and from the viewpoint of energy saving, iron loss should be as low as possible. Magnetostriction refers to a slight change in the external shape of an electrical steel sheet due to a change in the strength of magnetization when the electrical steel sheet is excited with an alternating current, and it is expected that this will cause the electrical steel sheet to vibrate and lead to noise. In order to control these magnetic properties, it is known that it is important to control the crystal orientation and magnetic domain structure in the electrical steel sheet. Therefore, analytical methods and devices for analyzing the crystal orientation and magnetic domain structure are required. In particular, grain-oriented electrical steel sheets develop a texture called the GOSS orientation, but if they deviate from the GOSS orientation, lancet magnetic domains are generated depending on the deviation angle, which deteriorates the magnetic properties, so detailed knowledge about the relationship between the crystal orientation and the magnetic domain structure is required.

Patent Document 1 uses the so-called Bitter method as a method for observing the magnetic domain structure. The Bitter method is a method in which a liquid containing magnetic colloidal particles is applied to a magnetic material, dried, and then the particle positions are observed. This method makes use of the property of magnetic colloidal particles gathering at magnetic domain walls, and by observing the particle positions after drying, the magnetic domain wall positions can be determined, providing insight into the size of the magnetic domains.

Patent Document 2 proposes a method for observing magnetic domains using a Kerr effect microscope that uses the magneto-optical Kerr effect as a method for determining not only the size of magnetic domains but also the magnetization direction within the magnetic domains. While a Kerr effect microscope can observe the in-plane two-dimensional magnetization direction of the observation surface, it cannot obtain information regarding the crystal orientation.

Patent Document 3 discloses a method of irradiating an electron beam onto a tilted sample to obtain a backscattered electron image, and observing the magnetization component parallel to the direction perpendicular to the tilt axis as magnetic domain contrast. This method makes it possible to observe magnetic domains by using an electron beam, but, like Patent Document 2, it is not possible to obtain information about crystal orientation at the same time.

In response to this, Non-Patent Document 1 discloses a method of obtaining information about crystal orientation from an Electron Backscatter Diffraction (EBSD) image obtained by tilting a sample inside a scanning electron microscope and irradiating it with an electron beam, and then obtaining magnetic domain contrast by constructing a reflected electron image from the image brightness of the EBSD image. Using this method, it is possible to correlate the distribution of magnetic domains in the observation area with the crystal orientation. However, the technology in Non-Patent Document 1 has the problem of not being able to obtain information about the magnetization direction within the magnetic domains.

In addition, backscattered electron images and secondary electron images obtained with a scanning electron microscope generally contain contrast due to unevenness on the sample surface. As a result, the contrast due to unevenness is superimposed on the contrast due to magnetic domains, making the contrast due to magnetic domains difficult to distinguish. To prevent this, it is necessary to prepare the sample surface so that it is flat and free of surface damage caused by sample preparation, but in reality, some unevenness on the sample surface is unavoidable.

Japanese Patent Application Publication No. 11-211805 JP 2010-203989 A JP 2010-92646 A

Gallaugher et al. Ultramicroscopy 142 (2014) 40

As described above, it has conventionally not been easy to obtain highly accurate information regarding the magnetic domain structure (the size of the magnetic domain and the magnetization direction within the magnetic domain) in a steel sheet.
An object of the present invention is to easily obtain highly accurate information on the magnetic domain structure in steel material.

The steel plate is made of iron having a body-centered cubic (BCC) structure, and the property that the magnetization direction is parallel or anti-parallel to the <100> direction is utilized, and by adjusting the surface of the sample to the (001) plane, the candidates for the magnetization direction in the analysis surface are limited to four directions, namely, the [100], [-100], [010], and [0-10] directions. As a result of intensive studies by the present inventors, it has been found that a method can be constructed for determining the magnetization direction by changing the magnetic domain contrast in the backscattered electron image by constructing a backscattered electron image from pixel intensities in a specific region of an EBSD image.
The gist of the present invention is as follows.
[1]
A method for observing magnetic domains in a steel sheet, the method comprising: irradiating a measurement point of a sample made of a steel sheet having a body centered cubic (BCC) structure with an electron beam in an electron microscope; and using an electron backscatter diffraction (EBSD) image showing a diffraction pattern of reflected electrons that reach an EBSD detector, the method comprising the steps of:
In the EBSD detector, when the electron beam is projected onto the EBSD detector, a direction parallel to the irradiation direction of the projected electron beam is defined as an irradiation parallel direction, a direction anti-parallel to the irradiation parallel direction is defined as an irradiation anti-parallel direction, and two directions perpendicular to the irradiation parallel direction and anti-parallel to each other are defined as a first irradiation side direction and a second irradiation side direction, respectively;
a sample preparation step of preparing the sample so that the sample surface corresponds to the (001) plane of the BCC structure;
a sample setting step of setting the sample in the electron microscope such that the [100] direction and the [−100] direction of the crystal correspond to one of the irradiation parallel direction and the irradiation anti-parallel direction, and the first irradiation side direction and the second irradiation side direction, and the [010] direction and the [0−10] direction of the crystal correspond to the other;
a first backscattered electron image creating step of calculating an intensity of backscattered electrons in a partial region of the EBSD image for the sample set in the sample setting step while varying the measurement point, thereby creating a backscattered electron image in the partial region of the EBSD image;
a second backscattered electron image creating step of creating a backscattered electron image based on the backscattered electron image created in the first backscattered electron image creating step, in which the contrast derived from magnetic domains having magnetization directions corresponding to the [100] direction and the [−100] direction and the contrast derived from magnetic domains having magnetization directions corresponding to the [010] direction and the [0-10] direction are emphasized and the contrast derived from the shape of the sample surface is alleviated compared to the backscattered electron image created in the first backscattered electron image creating step;
a magnetic domain structure acquiring step of acquiring information indicating the size, shape, and magnetization direction of each magnetic domain in the sample using the backscattered electron image created in the second backscattered electron image creating step;
having
The first reflected electron image creation step creates at least one of an anti-parallel-irradiation direction side reflected electron image which is a reflected electron image in at least a partial region of the EBSD image on the anti-parallel-irradiation direction side of the position corresponding to the measurement point and a parallel-irradiation direction side reflected electron image which is a reflected electron image in at least a partial region of the EBSD image on the parallel-irradiation direction side of the position corresponding to the measurement point, a first irradiation lateral direction side reflected electron image which is a reflected electron image in at least a partial region of the EBSD image on the first irradiation lateral direction side of the position corresponding to the measurement point, and a second irradiation lateral direction side reflected electron image which is a reflected electron image in at least a partial region of the EBSD image on the second irradiation lateral direction side of the position corresponding to the measurement point.
[2]
The method for observing magnetic domains in steel sheet described in [1], wherein in the first reflected electron image creation process, a reflected electron image on the anti-parallel direction side of the irradiation, a reflected electron image on the parallel direction side of the irradiation, a reflected electron image on the first side direction of the irradiation, and a reflected electron image on the second side direction of the irradiation are created.
[3]
The second backscattered electron image creation step performs addition of pixel values of the backscattered electron image on the anti-parallel direction side to pixel values of the backscattered electron image on the parallel direction side to create a first irradiation direction magnetic domain contrast-enhanced image, which is a backscattered electron image in which the contrast derived from the magnetic domain having a magnetization direction corresponding to one of the [100] direction and the [-100] direction, and the [010] direction and the [0-10] direction is enhanced and the contrast derived from the shape of the sample surface is relaxed; and performs subtraction of pixel values of the backscattered electron image on the first irradiation side direction side from pixel values of the backscattered electron image on the second irradiation side direction side to create an irradiation side direction magnetic domain contrast-enhanced image, which is a backscattered electron image in which the contrast derived from the magnetic domain having a magnetization direction corresponding to the other direction is enhanced and the contrast derived from the shape of the sample surface is relaxed.
[4]
The method for observing magnetic domains in a steel sheet according to [1] , wherein in the first reflected electron image creation process, a reflected electron image in the anti-parallel direction of the irradiation, a reflected electron image in the first side direction of the irradiation, and a reflected electron image in the second side direction of the irradiation are created, and a reflected electron image in the parallel direction of the irradiation is not created.
[5]
In the second backscattered electron image creation step, a pixel value of the backscattered electron image in the first irradiation side direction is added to a pixel value of the backscattered electron image in the second irradiation side direction, and then a pixel value of the backscattered electron image in the anti-parallel direction is subtracted to create a second irradiation direction magnetic domain contrast-enhanced image, which is a backscattered electron image in which the contrast derived from the magnetic domain having a magnetization direction corresponding to one of the [100] direction and the [-100] direction, and the [010] direction and the [0-10] direction is enhanced and the contrast derived from the shape of the sample surface is relaxed; and a pixel value of the backscattered electron image in the first irradiation side direction is subtracted from a pixel value of the backscattered electron image in the second irradiation side direction, which creates an irradiation side direction magnetic domain contrast-enhanced image, which is a backscattered electron image in which the contrast derived from the magnetic domain having a magnetization direction corresponding to the other direction is enhanced and the contrast derived from the shape of the sample surface is relaxed.
[6]
The method for observing magnetic domains in steel sheet according to any one of [1] to [5], wherein in the first reflected electron image creating step, the EBSD image is divided into a plurality of divided regions such that at least three regions exist in each of the directions corresponding to the irradiation parallel direction and the irradiation anti-parallel direction, and the directions corresponding to the first irradiation side direction and the second irradiation side direction, and a reflected electron image is created in each of the divided regions by using a pixel value of the divided region closest to an edge of the EBSD image, without using a pixel value of the divided region closest to a position corresponding to the measurement point in the EBSD image.
[7]
the backscattered electron image created in the first backscattered electron image creating step is an image including an observation region of a crystal orientation of the sample that is analyzed based on the EBSD image,
The method for observing magnetic domains in a steel sheet according to any one of [1] to [6], wherein the information indicating the size, shape, and magnetization direction of each magnetic domain in the sample is information of a region including an observation region of the crystal orientation of the sample.

The present invention makes it easy to obtain highly accurate information regarding the magnetic domain structure in steel material.

FIG. 1 illustrates an example of a configuration of an analysis system. 1 is a flowchart illustrating an example of a method for observing magnetic domains in a steel sheet. FIG. 1 is a diagram illustrating an example of an outline of processing in an information processing device. 1 is a diagram conceptually showing the difference in travel distance within a sample between reflected electrons traveling upward and reflected electrons traveling downward. FIG. FIG. 1 is a diagram conceptually illustrating how the unevenness of a sample surface reduces the number of reflected electrons traveling downward. 1 is a diagram conceptually explaining an example of the behavior of reflected electrons BE when a measurement point is a magnetic domain whose magnetization directions are +X direction and −X direction. 10 is a diagram conceptually explaining an example of the behavior of reflected electrons BE when the measurement points are magnetic domains whose magnetization directions are in the +Y direction and −Y direction. 10 is a diagram conceptually showing an example of a change in contrast caused by adding a pixel value of an upper reflected electron image and a pixel value of a lower reflected electron image. FIG. 10 is a diagram conceptually showing an example of a change in contrast caused by subtracting a pixel value of a left-side reflected electron image from a pixel value of a right-side reflected electron image. FIG. 13 is a diagram conceptually showing an example of a change in contrast caused by adding a pixel value of a left-side reflected electron image and a pixel value of a right-side reflected electron image and then subtracting a pixel value of an upper reflected electron image. FIG. FIG. 13 is a diagram showing an example of a crystal orientation map. FIG. 2 is a diagram showing an example of divided regions of an EBSD detector. FIG. 1 is a diagram (photograph) showing an example of an upper reflected electron image. FIG. 1 is a diagram (photograph) showing an example of a bottom reflected electron image. FIG. 13 is a diagram (photograph) showing an example of a first X-direction magnetic domain contrast enhanced image. FIG. 13 is a diagram (photograph) showing an example of a backscattered electron image to be compared with a first X-direction magnetic domain contrast enhanced image. FIG. 13 is a diagram (photograph) showing an example of a left-side reflected electron image. FIG. 13 is a diagram (photograph) showing an example of a right-side reflected electron image. FIG. 13 is a diagram (photograph) showing an example of a Y-direction magnetic domain contrast enhanced image. FIG. 13 is a diagram (photograph) showing an example of a Y-direction magnetic domain contrast relaxation image. FIG. 13 is a diagram (photograph) showing an example of a second X-direction magnetic domain contrast enhanced image. FIG. 2 is a diagram showing an example of a magnetic domain structure map. FIG. 1 is a diagram (photograph) showing an example of a full-area reflected electron image.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
In addition, the fact that the objects to be compared are the same in terms of length, position, size, interval, etc. includes not only the case where they are exactly the same, but also the case where they are different within the scope of the invention (for example, the difference is within the tolerance range determined at the time of design). Also, the symbol with a ● in a circle in each figure indicates the test from the back to the front. For example, if a symbol with a ● in a circle is placed next to +X, it indicates that the +X direction is the direction from the back to the front of the paper.

Figure 1 shows an example of the configuration of an analysis system that performs EBSD analysis. Figure 2 is a flowchart that explains an example of a method for observing magnetic domains in a steel sheet.

In FIG. 1, the analysis system is a system for observing (analyzing) the magnetic domains and magnetization direction in a steel sheet having a BCC structure. In this embodiment, the steel sheet is a bidirectional electromagnetic steel sheet, but the steel sheet is not limited to a bidirectional electromagnetic steel sheet, and may be, for example, a unidirectional electromagnetic steel sheet or a non-directional electromagnetic steel sheet, or may not be an electromagnetic steel sheet. The analysis system includes a scanning electron microscope 110, an EBSD detector 120, an imaging device 130, and an information processing device 140.

When the sample S is a steel plate having a BCC structure, the easy axis of magnetization of the iron is parallel to the principal axis of the crystal, and the magnetization direction of one magnetic domain in the iron is limited to one of the directions [100], [-100], [010], [0-10], [001], and [00-1]. Furthermore, if the surface of the sample S is limited to the (001) plane, the magnetization directions observable from the surface of the sample S are limited to only the directions [100], [-100], [010], and [0-10]. Therefore, the sample S is adjusted so that the surface of the sample S corresponds to the (001) plane (sample adjustment process of step S201). In order to make the surface of the sample S correspond to the (001) plane, the crystal orientation of the sample S to be observed is measured in advance by the EBSD method or the Laue method using X-rays, and the surface of the sample S is prepared (processed) so that the (001) plane becomes the surface of the sample S. If there is processing distortion on the surface of the sample S, the original magnetic domain structure will change, so it is preferable to perform a finishing process to remove the surface processing distortion using electrolytic polishing or the like. In addition, it is preferable to keep the deviation angle from the (001) plane to about ±5°. If the deviation angle is large, a leakage magnetic field may be generated on the surface of the sample S, which may significantly change the magnetic domain structure on the surface of the sample S. In this way, the surface of the sample S is not only perfectly parallel to the (001) plane, but also allows such deviation. The surface of the sample S corresponding to the (001) plane means that not only the surface of the sample S is perfectly parallel to the (001) plane, but also that such deviation is acceptable.

EBSD analysis is performed in a scanning electron microscope 110. The electron beam probe EP is focused and scanned by the optical system 111 in the scanning electron microscope. The optical system 111 in the scanning electron microscope includes an electron gun, a focusing lens, a scanning deflector, and the like. The optical system 111 in the scanning electron microscope is not particularly limited as long as it is installed in a commercially available known scanning electron microscope. The sample S is set in the sample holder 112 so that its surface (the surface irradiated with the electron beam probe EP) faces the optical system 111 and the EBSD detector 120 and is inclined with respect to the electron beam probe EP generated from the optical system 111 (sample setting process of step S202). FIG. 1 illustrates an example in which the inclination angle θ is an inclination angle with respect to the horizontal plane (the installation surface of the scanning electron microscope 110). In the EBSD method, an inclination angle θ of 70° is generally used to obtain a clear diffraction pattern (Kikuchi pattern), but a value of about 40 to 80° is also acceptable.

1 illustrates an example in which the irradiation direction of the electron beam probe EP is vertically downward. Here, when the electron beam probe EP is projected onto the EBSD detector 120 (the detection surface of the reflected electron BE), the direction parallel to the irradiation direction of the projected electron beam probe EP is the irradiation parallel direction, and the direction anti-parallel to the irradiation parallel direction is the irradiation anti-parallel direction. That is, the irradiation parallel direction is, for example, the direction of an orthogonal projection vector obtained by orthogonally projecting the electron beam probe EP onto the EBSD detector 120 (the detection surface of the reflected electron BE) as a vector pointing in the traveling direction of the electron beam probe EP. In addition, two directions perpendicular to the irradiation parallel direction and anti-parallel to each other are respectively the first irradiation side direction and the second irradiation side direction. FIG. 1 illustrates an example in which the EBSD detector 120 is arranged so that the EBSD detector 120 is parallel to the electron beam probe EP. Also, a case where the EBSD detector 120 is arranged along the up-down direction (vertical up-down direction) is illustrated. Therefore, projecting the electron beam probe EP onto the EBSD detector 120 corresponds to translating the electron beam probe EP to the position of the EBSD detector 120. Therefore, the direction parallel to the irradiation direction of the electron beam probe EP projected onto the EBSD detector 120 (irradiation parallel direction) is the downward direction (vertical downward direction), the irradiation anti-parallel direction is the upward direction (vertical upward direction), and the first irradiation side direction and the second irradiation side direction are left-right directions. More specifically, in this embodiment, the irradiation anti-parallel direction (upward direction) is the top, and the left direction when looking at the sample S side through the EBSD detector 120 is the first irradiation side direction, and the right direction is the second irradiation side direction. In FIG. 1, the downward direction (vertical downward direction) is the direction of travel of the electron beam probe EP (the direction toward the tip of the > attached to the straight line representing the electron beam probe EP), and the upward direction (vertical upward direction) is the opposite direction. Additionally, the direction from the front side to the back side of the paper in FIG. 1 is the left side, and the direction from the back side to the front side is the right side. In the following explanation, up, down, left, and right are assumed to be up, down, left, and right expressed as above.

Note that even if the direction parallel to the irradiation direction of the electron beam probe EP on the EBSD detector 120 (electron beam detection surface) when the EBSD detector 120 (electron beam detection surface) is assumed to be parallel to the electron beam probe EP, and the irradiation parallel direction, anti-parallel irradiation direction, first irradiation side direction, and second irradiation side direction are defined as the irradiation parallel direction, the irradiation parallel direction, anti-parallel irradiation direction, first irradiation side direction, and second irradiation side direction are the same directions as those expressed using the projection of the electron beam probe EP described above. In this way, when the irradiation parallel direction, anti-parallel irradiation direction, first irradiation side direction, and second irradiation side direction are defined, if the EBSD detector 120 is not parallel to the electron beam probe EP, the EBSD detector 120 changes the inclination angle with respect to the electron beam probe EP so that the EBSD detector 120 is parallel to the electron beam probe EP, and the irradiation parallel direction, anti-parallel irradiation direction, first irradiation side direction, and second irradiation side direction are determined as described above.

In addition, the irradiation direction of the electron beam probe EB is not limited to vertically downward. The parallel irradiation direction, anti-parallel irradiation direction, first irradiation side direction, and second irradiation side direction are each determined according to the irradiation direction of the electron beam probe EB in accordance with the definitions described above.

In the following description, to simplify the notation, when the electron beam probe EP is projected onto the surface of the sample S, the direction parallel to the irradiation direction of the projected electron beam probe EP is referred to as the -Y direction. That is, the -Y direction is, for example, the direction of an orthogonal projection vector obtained by orthogonally projecting the electron beam probe EP onto the surface of the sample S, regarding the electron beam probe EP as a vector pointing in the traveling direction of the electron beam probe EP. Also, the direction anti-parallel to the parallel irradiation direction on the surface of the sample S is referred to as the +Y direction. Also, of the two directions perpendicular to the parallel irradiation direction on the surface of the sample S and anti-parallel to each other, the +Y direction is considered to be up, and the leftward and rightward directions when the surface of the sample S is viewed from the EBSD detector 120 side are referred to as the -X direction and the +X direction, respectively.

In the example shown in FIG. 1, the upward direction corresponds to the +Y direction, the downward direction corresponds to the -Y direction, the right direction corresponds to the +X direction, and the left direction corresponds to the -X direction. Note that the upward direction corresponds to the +Y direction means that the upward direction is the +Y direction when the sample S and the EBSD detector 120 are arranged in parallel, and the downward direction corresponds to the -Y direction means that the downward direction is the -Y direction when the sample S and the EBSD detector 120 are arranged in parallel. Also, even if the ±X direction and ±Y direction are expressed with the direction parallel to the irradiation direction of the electron beam probe EP on the surface of the sample S in a state in which the surface of the sample S is parallel to the electron beam probe EP as the -Y direction, the ±X direction and ±Y direction are the same directions as those expressed using the projection of the electron beam probe EP described above.

As described above, the crystal orientation of the sample S to be observed is measured in advance by the EBSD method or the Laue method using X-rays, so that the direction of the principal axis of the crystal of the sample S can be easily identified. In this embodiment, the [100] and [-100] directions of the sample S correspond to the +X and -X directions of the sample S surface, respectively, and the [010] and [0-10] directions of the sample S correspond to the +Y and -Y directions of the sample S surface, respectively. It is preferable to suppress the deviation angle between the [100] and [010] directions and the +X and +Y directions of the sample S to about ±5°. In this way, not only are the directions of the crystal axes of the sample S ([100], [-100], [010], [0-10] directions) completely parallel to the ±X and ±Y directions, but also allow such deviation. The crystal axis directions ([100], [-100], [010], [0-10] directions) of the sample S corresponding to the ±X direction and ±Y direction means that the crystal axis directions of the sample S are not only completely parallel to the ±X direction and ±Y direction, but also allow such deviation. The [100] and [-100] directions of the sample S may correspond to the -X direction and +X direction of the sample S surface, respectively, or the [010] and [0-10] directions may correspond to the -Y direction and +Y direction of the sample S surface, respectively. The [100] and [-100] directions may correspond to either the -Y direction or +Y direction of the sample S surface, or the [010] and [0-10] directions may correspond to either the -X direction or +X direction of the sample S surface, respectively.

In this manner, the electron beam probe EP is focused and scanned on the sample S set in the sample holder 112. The optical system in the scanning electron microscope is composed of an electron gun, focusing lens, scanning deflector, etc., and is not particularly limited as long as it is one that is installed in a commercially available known scanning electron microscope.

When an electron beam probe EB is irradiated onto a measurement point on a tilted sample S, backscattered electrons BE flying out from the surface of the sample S toward the EBSD detector 120 reach the EBSD detector 120. The EBSD detector 120 includes, for example, a fluorescent screen as a detection surface for the backscattered electrons. When the backscattered electrons that reach the EBSD detector 120 hit the EBSD detector 120 (fluorescent screen), light is generated. The generated light reaches an image pickup element (for example, a CCD (Charge Coupled Device) detector or a CMOS (Complementary MOS) detector via a condenser lens in the image pickup device 130 and is converted into an electric signal. The electric signal is stored as an EBSD image, which is an image showing a diffraction pattern of the backscattered electrons that reach the EBSD detector 120, with a resolution defined by the number of pixels of the image pickup element. The EBSD image has, as a pixel value, an electric signal intensity normalized for each pixel. For example, when the pixel value of each pixel is stored as 8-bit data, the pixel value is expressed as a luminance value of 256 gradations.
The scanning electron microscope 110, the EBSD detector 120, and the imaging device 130 themselves may be realized by known devices and are not limited to the configurations described above.

The information processing device 140 is a device that executes information processing to acquire information on the crystal orientation and magnetic domain structure (size, shape, and magnetization direction of each magnetic domain) in the steel plate sample S based on the EBSD image. The information processing device 140 includes an image acquisition unit 141, a crystal orientation acquisition unit 142, a first backscattered electron image creation unit 143, a second backscattered electron image creation unit 144, and a magnetic domain structure acquisition unit 145 as its functions. The hardware of the information processing device 140 is not particularly limited, and is realized, for example, by a computer equipped with a central processing unit, a main memory device, an auxiliary memory device, an input device, and an output device. FIG. 3 is a diagram showing an example of an overview of the processing in the information processing device 140.

The image acquisition unit 141 acquires an EBSD image 310 as shown in FIG. 3 (EBSD image acquisition process of step S203). As shown in FIG. 3, in this embodiment, the image acquisition unit 141 acquires an EBSD image 310 at each of a plurality of measurement points MP set in the observation area of the sample S. In this embodiment, the observation area of the crystal orientation of the sample S and the observation area of the magnetic domain structure of the sample S are the same, and the measurement points MP for observing the crystal orientation of the sample S and the measurement points MP for observing the magnetic domain structure of the sample S are the same. However, the measurement points for observing the crystal orientation of the sample S and the measurement points for observing the magnetic domain structure of the sample S may be different. For example, the magnetic domain structure of the sample S may be wider than the observation area of the crystal orientation of the sample S, and the number of measurement points for observing the magnetic domain structure of the sample S may be greater than the number of measurement points for observing the crystal orientation of the sample S.

The crystal orientation acquisition unit 142 acquires information indicating the crystal orientation in the sample S based on the EBSD image 310 at each measurement point MP of the sample S (crystal orientation acquisition process of step S204). FIG. 3 illustrates a case where the information indicating the crystal orientation in the sample S is a crystal orientation map 320. The crystal orientation map 320 is an image created by storing, for each measurement point MP, a pixel value (color) indicating a value corresponding to the crystal orientation calculated from the EBSD image 310 obtained by irradiating the measurement point MP of the sample S with an electron beam probe EP in a region (pixel) 321 corresponding to the measurement point MP. The analysis of the crystal orientation is realized by EBSD analysis (for example, analysis of the position of the Kikuchi band, etc.) and is realized by a known technology, so a detailed description thereof will be omitted here. In addition, the information processing device 140 may acquire the residual stress (residual strain) in the sample S based on, for example, the image shift between multiple EBSD images 310 at each measurement point MP of the sample S. The method for analyzing the residual stress (residual strain) of S in the sample can be achieved using well-known technology, so a detailed explanation will be omitted here.

The first reflected electron image creation unit 143 creates a reflected electron image 330 based on the EBSD image 310 at each measurement point MP of the sample S (first reflected electron image creation process of step S205 in FIG. 2). The reflected electron image 330 is an image created by storing, for each measurement point MP, a pixel value indicating the intensity of the reflected electrons calculated from the EBSD image 310 obtained by irradiating the measurement point MP of the sample S with the electron beam probe EP in an area (pixel) 331 corresponding to the measurement point MP. As described above, the EBSD image 310 stores the normalized electrical signal intensity for each pixel as a pixel value. Therefore, by accumulating the pixel values, a value proportional to the number of reflected electrons (intensity of reflected electrons) that have reached the EBSD detector 120 can be calculated. Therefore, in this embodiment, a case in which the number of reflected electrons (intensity of reflected electrons) is expressed by an accumulated value of the pixel values of the EBSD image 310 is illustrated.

However, in this embodiment, the first reflected electron image creation unit 143 does not calculate the integrated value of the pixel values of all the regions of the EBSD image 310, but calculates the integrated value of a part of the region of the EBSD image 310 to create a reflected electron image. Specifically, in this embodiment, the first reflected electron image creation unit 143 integrates pixel values of a region above the position corresponding to the measurement point MP to create an upper reflected electron image as an example of an anti-parallel direction side reflected electron image, integrates pixel values of a region below the position corresponding to the measurement point MP to create a lower reflected electron image as an example of an parallel direction side reflected electron image, integrates pixel values of a region to the left of the position corresponding to the measurement point MP to create a left reflected electron image as an example of a first lateral direction side reflected electron image, and integrates pixel values of a region to the right of the position corresponding to the measurement point MP to create a reflected electron image including a right reflected electron image as an example of a second lateral direction side reflected electron image. However, as described below, the first reflected electron image creation unit 143 may create at least one of the upper reflected electron image and the lower reflected electron image. Here, the position of the region of the EBSD image 310 corresponding to the measurement point MP is the coordinate position in the EBSD image 310 corresponding to the foot of a perpendicular line drawn from the intersection (measurement point MP) between the electron beam probe EP and the sample S to the EBSD detector 120. In the example shown in FIG. 1, the coordinate positions of the measurement point MP in the up, down, left, and right directions are the same as the coordinate positions of the EBSD detector 120 in the up, down, left, and right directions.

The upper reflected electron image is obtained by calculating the intensity of reflected electrons in at least a portion of the region of the EBSD image 310 above the position corresponding to the measurement point MP, with the measurement point MP being varied. In this embodiment, the position corresponding to the measurement point MP is the center of the EBSD image 310. Therefore, the upper reflected electron image is an image in which the integrated value of at least a portion of the pixel values in the upper half of the EBSD image 310 obtained by irradiating the measurement point MP with the electron beam probe EP is stored in the region (pixel) 331 corresponding to the measurement point MP.

The left-side reflected electron image is obtained by calculating the intensity of reflected electrons in at least a portion of the region of the EBSD image 310 to the left of the position corresponding to the measurement point MP, with the measurement point MP being changed. Specifically, the left-side reflected electron image is an image in which the integrated value of at least a portion of the pixel values in the left half of the EBSD image 310 obtained by irradiating the measurement point MP with the electron beam probe EP is stored in the region (pixel) 331 corresponding to the measurement point MP.

The right-side reflected electron image is obtained by calculating the intensity of reflected electrons in at least a portion of the region of the EBSD image 310 to the right of the position corresponding to the measurement point MP, with the measurement point MP being changed. Specifically, the right-side reflected electron image is an image in which the integrated value of at least a portion of the pixel values in the right half of the EBSD image 310 obtained by irradiating the measurement point MP with the electron beam probe EP is stored in the region (pixel) 331 corresponding to the measurement point MP.

The lower backscattered electron image is obtained by calculating the intensity of backscattered electrons in at least a portion of the region of the EBSD image 310 below the position corresponding to the measurement point MP, with the measurement point MP being varied. Specifically, the lower backscattered electron image is an image in which the integrated value of at least a portion of the pixel values in the lower half of the EBSD image 310 obtained by irradiating the measurement point MP with the electron beam probe EP is stored in the region (pixel) 331 corresponding to the measurement point MP.

The second backscattered electron image creation unit 144 creates a backscattered electron image based on the backscattered electron images (upper backscattered electron image, lower backscattered electron image, left backscattered electron image, and right backscattered electron image) created by the first backscattered electron image creation unit 143 in which the contrast resulting from the magnetic domains having magnetization directions corresponding to the [100] direction and the [-100] direction, or the contrast resulting from the magnetic domains having magnetization directions corresponding to the [010] direction and the [0-10] direction is emphasized (increased) and the contrast resulting from the shape (irregularities) of the surface of the sample S is alleviated (reduced) compared to the backscattered electron images (second backscattered electron image creation process of step S206 in FIG. 2).

The contrast observed in the backscattered electron image does not only originate from the magnetic domain structure (magnetic domains and magnetization direction), but also from the shape (convex/concave) of the surface of the sample S. This makes it difficult to obtain contrast originating from the true magnetic domain structure (magnetic domains and magnetization direction). This is particularly problematic when quantifying the magnitude of magnetization from the magnitude of contrast originating from the magnetic domain structure (magnetic domains and magnetization direction). In the following explanation, the contrast originating from the shape of the surface of the sample S is also referred to as the convex/concave contrast. The contrast originating from the magnetic domain structure (magnetic domains and magnetization direction) is also referred to as the magnetic domain contrast.

To prevent the above-mentioned topography contrast from being superimposed on the magnetic domain contrast, it is important to minimize the topography formed during surface finishing of the sample S, but it is extremely difficult to achieve a surface finish without any topography. Therefore, the inventors thought that if it were possible to reduce the topography contrast that is unavoidably present in the backscattered electron image, it would be possible to enhance the magnetic domain contrast.

As a result, the inventors found that the contrast appearing in the upper backscattered electron image is mainly expressed as topographical contrast. There are two possible reasons for this: Figure 4A is a diagram explaining why the contrast appearing in the upper backscattered electron image is mainly expressed as topographical contrast. Specifically, Figure 4A is a diagram conceptually illustrating the difference in the travel distance in the sample between backscattered electrons heading upward and backscattered electrons heading downward.

Next, the first reason will be explained.
The Lorentz force F, which is a force acting on a charge (here, an electron) when a charge q moves with a velocity v in a space where a magnetic field of magnetic flux density B exists, is expressed as F=qvBsinθ. θ is the angle between the vector of magnetic flux density B and the vector of velocity v. In order to increase the magnetic domain contrast, it is necessary to increase the Lorentz force F so that the trajectory of the electron changes significantly due to the influence of the magnetic field in the sample S. In this embodiment, since the case where the electron beam probe EP is directed downward is illustrated, as shown in FIG. 4A, the moving distance of the reflected electrons in the sample S from the inclined sample S toward the surface of the sample S is shorter in the case where the reflected electrons are directed upward than in the case where the reflected electrons are directed downward. Therefore, the distance over which the reflected electrons BE are subjected to the Lorentz force F is smaller in the case where the reflected electrons are directed upward than in the case where the reflected electrons are directed downward. Therefore, the magnetic domain contrast is easily alleviated in the upper reflected electron image. Furthermore, in the surface region of the sample S without unevenness, the lower reflected electron image is easily brighter than the upper reflected electron image. This is because the electric wire probe EP faces downward, and thus reflected electrons tend to be generated downward.

In addition, since the Lorentz force F acting on the incident electrons and the reflected electrons can be increased by increasing the speed of the incident electrons entering the sample S, it is preferable that the acceleration voltage of the electron gun is large in the EBSD image acquisition process of step S203 in FIG. 2. For example, it is preferable to set the acceleration voltage of the electron gun to 20 kV or more. If the acceleration voltage of the electron gun is less than 20 kV, the Lorentz force F acting on the reflected electrons is small, and it may be difficult to generate magnetic domain contrast in the reflected electron image appearing on the EBSD detector 120.

The second reason is that in this embodiment, the electron beam probe EP is directed downward, so the energy of the reflected electrons directed upward is smaller than the energy of the reflected electrons directed downward. Therefore, the magnetic domain contrast is likely to be reduced in the upper reflected electron image.

From the above, it can be considered that the contrast appearing in the upper backscattered electron image is dominated by topography contrast rather than magnetic domain contrast, and that the contrast appearing in the upper backscattered electron image can be regarded as a contrast in which the shape of the surface of the sample S is emphasized.

The inventors have also found that the topography contrast is reversed between the upper backscattered electron image and the lower backscattered electron image. The reason for this will be described with reference to FIG. 4B. FIG. 4B is a diagram for explaining the topography contrast being reversed between the side backscattered electron image and the lower backscattered electron image. Specifically, FIG. 4B is a diagram conceptually showing that the amount of backscattered electrons moving downward is reduced due to the unevenness of the surface of the sample S. In FIG. 4B, when the backscattered electrons traveling from the illustrated backscattered electron generation point toward the upper side reach a position above the generation point in the region of the EBSD detector 120, there is nothing blocking the backscattered electrons and they are observed brightly. On the other hand, when the backscattered electrons traveling from the illustrated backscattered electron generation point toward the lower side reach a position below the generation point in the region of the EBSD detector 120, they are blocked by the convex portion and observed darkly. Therefore, it is considered that the lower backscattered electron image becomes dark at a position where the upper backscattered electron image becomes bright (pixel value becomes large), and the lower backscattered electron image becomes bright at a position where the upper backscattered electron image becomes dark.

Based on the above, the inventors have found that by using at least one of the upper backscattered electron image and the lower backscattered electron image, it is possible to extract the contrast resulting from the shape (unevenness) occurring in the ±Y direction on the surface of the sample S as unevenness contrast. Below, a specific example of a method in which the second backscattered electron image creation unit 144 uses the upper backscattered electron image, the lower backscattered electron image, the left backscattered electron image, and the right backscattered electron image to extract the contrast resulting from the magnetic domains having magnetization directions corresponding to the [100] direction and the [-100] direction, and the contrast resulting from the magnetic domains having magnetization directions corresponding to the [010] direction and the [0-10] direction, is described.

First, the influence of the internal magnetic field of the sample S on the trajectory of the reflected electrons flying from the surface of the sample S toward the EBSD detector 120 when the tilted sample S is irradiated with the electron beam probe EP will be considered.
Fig. 5 is a diagram conceptually explaining an example of the behavior of the reflected electrons BE when the measurement point MP is a magnetic domain whose magnetization directions are the +X direction (direction corresponding to the [100] direction) and the -X direction (direction corresponding to the [-100] direction). Fig. 6 is a diagram conceptually explaining an example of the behavior of the reflected electrons BE when the measurement point MP is a magnetic domain whose magnetization directions are the +Y direction (direction corresponding to the [010] direction) and the -Y direction (direction corresponding to the [0-10] direction).

5A, when the internal magnetic field is magnetized in the +X direction in the sample S, the reflected electrons BE heading toward the EBSD detector 120 are subjected to a downward (downward) Lorentz force F inside the sample S. Therefore, the arrival position 511 of the reflected electrons BE on the detector 120 is shifted downward from the arrival position 510 in the absence of an internal magnetic field. Conversely, as shown in FIG. 5B, when the internal magnetic field is magnetized in the -X direction in the sample S, the reflected electrons BE heading toward the EBSD detector 120 are subjected to an upward (upward) Lorentz force F inside the sample S. Therefore, the arrival position 512 of the reflected electrons BE on the detector 120 is shifted upward from the arrival position 510 in the absence of an internal magnetic field. In this case, if a backscattered electron image is created using only the backscattered electrons BE that reach above the center of the EBSD detector 120, the magnetic domains whose magnetization direction is in the -X direction will be observed brightly, but for the reasons described above, in such a backscattered electron image, the magnetic domain contrast in the magnetization direction in the -X direction will be relaxed and the concave-convex contrast will be emphasized. On the other hand, if a backscattered electron image is created using only the backscattered electrons BE that reach below the center of the EBSD detector 120, the magnetic domains whose magnetization direction is in the +X direction will be observed brightly.

Similarly, when a backscattered electron image is created using only the backscattered electrons BE that have reached the right side of the center of the EBSD detector 120, the magnetic domains in the sample S whose magnetization direction is the +Y direction are observed brightly (see the arrival position 510 of the backscattered electrons BE when there is no internal magnetic field and the arrival position 611 of the backscattered electrons BE when the internal magnetic field is magnetized in the +Y direction, as shown in FIG. 6(a)). Also, when a backscattered electron image is created using only the backscattered electrons BE that have reached the left side of the center of the EBSD detector 120, the magnetic domains in the sample S whose magnetization direction is the -Y direction are observed brightly (see the arrival position 510 of the backscattered electrons BE when there is no internal magnetic field and the arrival position 612 of the backscattered electrons BE when the internal magnetic field is magnetized in the -Y direction, as shown in FIG. 6(b)).

As described above, in this embodiment, the [100] and [-100] directions of the sample S correspond to the +X and -X directions of the surface of the sample S, respectively. Therefore, for example, when a backscattered electron image (lower backscattered electron image) is created with backscattered electrons BE that have reached a position below the center of the EBSD detector 120, the brightest observed region corresponds to a magnetic domain having a magnetization direction in the +X direction (i.e., a direction corresponding to the [100] direction), and the darkest observed region corresponds to a magnetic domain having a magnetization direction in the -X direction (i.e., a direction corresponding to the [-100] direction). Furthermore, when a backscattered electron image (lower backscattered electron image) is created with backscattered electrons BE that have reached a position below the center of the EBSD detector 120, the topography contrast is inverted compared to when a backscattered electron image (upper backscattered electron image) is created with backscattered electrons BE that have reached a position above the center of the EBSD detector 120. Therefore, when a backscattered electron image (lower backscattered electron image) is created using backscattered electrons BE that have reached a position lower than the center of the EBSD detector 120, the topography contrast is superimposed on the magnetic domain contrast of the magnetization direction in the ±X direction. In addition, the [010] and [0-10] directions of the sample S correspond to the +Y and -Y directions on the surface of the sample S, respectively. Therefore, for example, when a backscattered electron image (left backscattered electron image) is created using backscattered electrons BE that have reached a position to the left of the center of the EBSD detector 120, the brightest observed region corresponds to a magnetic domain having a magnetization direction in the -Y direction (i.e., a direction corresponding to the [0-10] direction), and the darkest observed region corresponds to a magnetic domain having a magnetization direction in the +Y direction (i.e., a direction corresponding to the [010] direction).

As described above, when a backscattered electron image is created using backscattered electrons BE that reach the lower, left, and right sides of the center of the EBSD detector 120, magnetic domains having a specific magnetization direction can be displayed brightly (or darkly), and the contrast of the magnetic domains in that magnetization direction can be emphasized. In addition, when a backscattered electron image is created using backscattered electrons BE that reach the upper side of the center of the EBSD detector 120, the concave-convex contrast can be emphasized.

In this case, if a region close to the edge of the EBSD detector 120 is selected, the magnetic domain contrast and the topography contrast are likely to be emphasized. This is because, when the electron trajectory is changed by the Lorentz force F, the electrons moving toward the edge of the EBSD detector 120 show a larger change in the arrival position on the EBSD detector 120 than the electrons moving toward the center of the EBSD detector 120. In a normal EBSD measurement setup, it is preferable to integrate the pixel values of regions corresponding to an area of 1/20 or more of the area of the EBSD detector 120 (the detection area of the reflected electron image of the EBSD detector 120) on each of the upper, lower, left, and right sides of the center of the EBSD image 310. If the area to be integrated is small, the number of pixel value counts will decrease, and sufficient magnetic domain contrast and topography contrast may not be obtained. In addition, it is preferable not to select a region close to the center of the EBSD detector 120. For example, in a region close to the center of the EBSD detector 120, the influence of magnetic domains in the +X direction, -X direction, +Y direction, and -Y direction may be large. For example, the EBSD image 310 may be divided into a plurality of regions so that there are at least three regions in each of the vertical and horizontal directions, and an upper reflected electron image, a lower reflected electron image, a left reflected electron image, and a right reflected electron image may be created using pixel values of the divided region 310 located closest to the edge of the EBSD image, without using pixel values of at least the divided region located closest to the center of the EBSD image 301. The number of divisions of the EBSD image 310 is not limited to this, and may be divided into four, for example, vertically, horizontally, and vertically.

For example, if the EBSD detector 120 is divided into four regions vertically and horizontally, the magnetic domains in the -Y direction will be bright in the upper left and lower left divided regions in addition to the magnetic domains in the -X and +X directions. Also, in the lower left and lower right divided regions of the backscattered electron image, the magnetic domains in the +X direction will be bright in addition to the magnetic domains in the -Y and +Y directions. Therefore, it is not easy to accurately extract the magnetic domain structure (magnetic domains and magnetic domain direction) by simply adding the pixel values of the EBSD image 310 for each region above, below, left, and right of the center of the EBSD image 310 (by simply creating an upper backscattered electron image, a lower backscattered electron image, a left backscattered electron image, and a right backscattered electron image).

Therefore, in this embodiment, the second reflected electron image creation unit 144 performs subtraction between the pixel values of the left reflected electron image and the pixel values of the right reflected electron image. The second reflected electron image creation unit 144 also performs either of the following: adding the pixel values of the upper reflected electron image and the pixel values of the lower reflected electron image, or subtracting the pixel value of the upper reflected electron image from the sum of the pixel values of the left reflected electron image and the pixel values of the right reflected electron image. The addition and subtraction of pixel values are performed at the same pixel positions. The reason for performing such calculations is explained below.

As explained with reference to FIG. 4B, at measurement point MP where unevenness contrast occurs, if the upper reflected electron image is bright (if the pixel value is large), the lower reflected electron image is dark, and if the upper reflected electron image is dark, the lower reflected electron image is bright. From this, the inventors discovered that adding the pixel values of the upper and lower reflected electron images reduces the unevenness contrast and enhances the magnetic domain contrast in the ±X directions.

Figure 7 is a conceptual diagram showing an example of the change in contrast caused by adding pixel values of an upper reflected electron image and a lower reflected electron image. For simplicity of explanation, it is assumed here that in magnetic domains magnetized in the -X direction, only pixel values resulting from unevenness appear in the reflected electron image. Also, in magnetic domains magnetized in the +X direction, only pixel values resulting from magnetization in the +X direction and pixel values resulting from unevenness appear. Also, the pixel value resulting from magnetization in the +X direction is set to 10, and the pixel value resulting from unevenness is set to 1.

As shown in FIG. 7A, when an unevenness contrast occurs in which the pixel value of the magnetic domain 701 magnetized in the -X direction is larger than that of the magnetic domain 702 magnetized in the +X direction adjacent to the magnetic domain 701, the pixel value of the upper reflected electron image of the magnetic domain 701 is 1, and the pixel value of the upper reflected electron image of the magnetic domain 702 is 0. When only the unevenness of the surface of the sample S is taken into consideration, if the upper reflected electron image of the magnetic domain 701 is bright, the lower reflected electron image of the magnetic domain 701 is dark. Therefore, the pixel value of the lower reflected electron image of the magnetic domain 701 is 0. Conversely, when only the unevenness of the surface of the sample S is taken into consideration, if the upper reflected electron image of the magnetic domain 702 is dark, the lower reflected electron image of the magnetic domain 702 is bright. Therefore, the pixel value of the lower reflected electron image of the magnetic domain 702 is a value (=11) obtained by adding the pixel value (=10) resulting from magnetization in the +X direction to the pixel value (=1) resulting from the unevenness. Therefore, by adding the pixel values of the upper reflected electron image and the lower reflected electron image, the unevenness contrast contained in the lower reflected electron image can be mitigated, the original magnetic domain contrast in the ±X direction (=10 (=11-1)) is obtained, and the magnetic domain contrast in the ±X direction is emphasized.

7B, in the case where an unevenness contrast occurs in which the pixel value of the magnetic domain 702 magnetized in the +X direction is larger than that of the magnetic domain 701 magnetized in the -X direction, the pixel value of the upper reflected electron image of the magnetic domain 702 is 1, and the pixel value of the upper reflected electron image of the magnetic domain 701 is 0. When only the unevenness of the surface of the sample S is taken into consideration, if the upper reflected electron image of the magnetic domain 701 is dark, the lower reflected electron image of the magnetic domain 701 is bright. Therefore, the pixel value of the lower reflected electron image of the magnetic domain 701 is 1. Conversely, when only the unevenness of the surface of the sample S is taken into consideration, if the upper reflected electron image of the magnetic domain 702 is bright, the lower reflected electron image of the magnetic domain 702 is dark. Therefore, the pixel value of the lower reflected electron image of the magnetic domain 702 is only the pixel value (=10) resulting from magnetization in the +X direction. Therefore, by adding the pixel values of the upper reflected electron image and the lower reflected electron image, the unevenness contrast contained in the lower reflected electron image can be alleviated, the original magnetic domain contrast in the ±X directions (=10 (=11-1)) can be obtained, and the magnetic domain contrast in the ±X directions is emphasized.
In this embodiment, a backscattered electron image in which the magnetic domain contrast in the ±X directions is emphasized and the unevenness contrast is alleviated by adding up the pixel values of the upper backscattered electron image and the lower backscattered electron image is created as an example of a first irradiation direction magnetic domain contrast emphasized image. In the following description, this backscattered electron image is also referred to as a first X-direction magnetic domain contrast emphasized image.

The inventors also discovered that subtracting the pixel values of the left-side reflected electron image from the pixel values of the right-side reflected electron image can enhance the magnetic domain contrast in the ±Y directions and reduce the unevenness contrast.

Figure 8 is a conceptual diagram showing an example of the change in contrast caused by subtracting pixel values of the left-side reflected electron image from pixel values of the right-side reflected electron image. For simplicity's sake, it is assumed that in a magnetic domain magnetized in the +Y direction, only pixel values resulting from magnetization in the +Y direction and pixel values resulting from unevenness appear, and in a magnetic domain magnetized in the -Y direction, only pixel values resulting from magnetization in the -Y direction and pixel values resulting from unevenness appear. In addition, the pixel values resulting from magnetization in the ±Y directions are each set to 10, and the pixel value resulting from unevenness is set to 1.

As shown in FIG. 8(a), if there is no concavo-convex contrast between a magnetic domain 801 magnetized in the -Y direction and a magnetic domain 802 magnetized in the +Y direction adjacent to the magnetic domain 801, subtracting the pixel value of the right-side reflected electron image from the pixel value of the left-side reflected electron image will result in the magnetic domain contrast in the ±Y directions changing from 10 to 20 (= 10 - (-10)), thereby enhancing the magnetic domain contrast in the ±Y directions.

As shown in FIG. 8B, when a concave-convex contrast occurs in which the pixel value of the magnetic domain 801 magnetized in the -Y direction is larger than that of the magnetic domain 802 magnetized in the +Y direction, the pixel value derived from the concave-convex is larger for the magnetic domain 801 in both the left-side reflected electron image and the right-side reflected electron image. Therefore, the pixel values of the left-side reflected electron image and the right-side reflected electron image of the magnetic domain 801 are 11 and 1, respectively. Therefore, when the pixel value of the right-side reflected electron image is subtracted from the pixel value of the left-side reflected electron image, the magnetic domain contrast in the ±Y directions becomes 10 to 20 (= 10 - (-10)), which is the same value as the magnetic domain contrast in the ±Y directions when there is no concave-convex contrast as shown in FIG. 8A. In this way, the magnetic domain contrast in the ±Y directions can be enhanced and the concave-convex contrast can be mitigated.

As shown in Fig. 8C, when a concavo-convex contrast occurs in which the pixel value of the magnetic domain 802 magnetized in the +Y direction is larger than that of the magnetic domain 801 magnetized in the -Y direction, the pixel value derived from the concavo-convex is larger for the magnetic domain 802 in both the left-side reflected electron image and the right-side reflected electron image. Therefore, the pixel values of the left-side reflected electron image and the right-side reflected electron image of the magnetic domain 802 are 1 and 11, respectively. Therefore, when the pixel value of the right-side reflected electron image is subtracted from the pixel value of the left-side reflected electron image, the magnetic domain contrast in the ±Y directions becomes 10 to 20 (= 10 - (-10)), which is the same value as the magnetic domain contrast in the ±Y directions in the absence of the concavo-convex contrast shown in Fig. 8A. In this way, the magnetic domain contrast in the ±Y directions can be enhanced and the concavo-convex contrast can be alleviated.
In the following description, a backscattered electron image in which the magnetic domain contrast in the ±Y directions is emphasized and the unevenness contrast is alleviated by subtracting the pixel values of the upper reflected electron image from those of the lower reflected electron image is created as an example of an irradiation side direction magnetic domain contrast emphasized image. In the following description, the backscattered electron image is also referred to as a Y direction magnetic domain contrast emphasized image. The subtraction of the pixel values of the upper reflected electron image from those of the lower reflected electron image may be realized by subtracting the pixel values of the lower reflected electron image from those of the upper reflected electron image, or may be realized by subtracting the pixel values of the lower reflected electron image from those of the lower reflected electron image. Furthermore, since it is only necessary to know the contrast of the images (difference in pixel values (brightness values)), if the subtraction results in a negative value, the pixel values after the subtraction may be normalized (for example, expressed as a relative value with the minimum value set to 0).

The inventors have also found that adding the pixel values of the left-side backscattered electron image and the right-side backscattered electron image averages the magnetic domain contrast in the ±Y directions, and thus emphasizes the magnetic domain contrast in the ±X directions. For example, in the lower left or lower right region of the EBSD image 310, magnetic domains having a magnetization direction in the +X direction appear bright, and magnetic domains having a magnetization direction in the -X direction appear dark. Therefore, adding the pixel values of the left-side backscattered electron image and the right-side backscattered electron image emphasizes the magnetic domain contrast in the ±X directions. However, in this case, the concave-convex contrast is not alleviated, so the concave-convex contrast is alleviated by adding the pixel values of the left-side backscattered electron image and the right-side backscattered electron image and then subtracting the pixel value of the upper backscattered electron image.

Figure 9 is a conceptual diagram showing an example of the change in contrast caused by adding the pixel values of the left reflected electron image and the right reflected electron image and then subtracting the pixel values of the upper reflected electron image. Here, a case is illustrated in which a magnetic domain 901 magnetized in the -X direction contains a pixel value resulting from magnetization in the -Y direction, and a magnetic domain 902 magnetized in the +X direction adjacent to the magnetic domain 901 contains a pixel value resulting from magnetization in the +Y direction. Also, a case is illustrated in which an unevenness contrast occurs in which the magnetic domain 901 magnetized in the -X direction has a larger pixel value than the magnetic domain 902. The pixel value resulting from magnetization in the +X direction is 7, the pixel value resulting from magnetization in the ±Y directions is 3, and the pixel value resulting from unevenness is 1.

As shown in FIG. 9(a), when only the pixel values resulting from magnetization in the ±Y directions contained in magnetic domains 901 and 902 are extracted and the pixel values of the left-side reflected electron image and the right-side reflected electron image are added together, the magnetic domain contrast in the ±Y directions is relaxed from 3 (=3-0) to 0 (=3-3).

As shown in Fig. 9(b), when only the pixel values resulting from magnetization in the ±X directions contained in the magnetic domains 901, 902 are extracted and the pixel values of the left-side reflected electron image and the right-side reflected electron image are added, the magnetic domain contrast in the ±X directions changes from 6 (=7-1) to 12 (=14-2), and the magnetic domain contrast in the ±X directions is emphasized. Then, as shown in Fig. 9(c), when the pixel values of the upper reflected electron image are subtracted from the pixel values thus added, the magnetic domain contrast in the ±X directions changes from 12 to 13 (=14-1), and the concavo-convex contrast (=1) can be alleviated.
In this embodiment, a backscattered electron image in which the magnetic domain contrast in the ±Y directions is relaxed and the magnetic domain contrast in the ±X directions is enhanced is created by adding the pixel values of the left-side backscattered electron image and the right-side backscattered electron image. In the following description, this backscattered electron image is also referred to as a Y-direction magnetic domain contrast relaxed image. In this embodiment, a backscattered electron image in which the magnetic domain contrast in the ±X directions is enhanced and the unevenness contrast is relaxed is created as an example of a second irradiation direction magnetic domain contrast enhanced image by subtracting the pixel values of the upper backscattered electron image from the pixel values of the Y-direction magnetic domain contrast relaxed image. In the following description, this backscattered electron image is also referred to as a second X-direction magnetic domain contrast enhanced image. Note that the second backscattered electron image creation unit 144 does not necessarily need to create both the first X-direction magnetic domain contrast enhanced image and the second X-direction magnetic domain contrast enhanced image, and may create only one of them. When creating a first X-direction magnetic domain contrast enhanced image, an upper reflected electron image, a lower reflected electron image, a left reflected electron image, and a right reflected electron image are required, so the first reflected electron image creation unit 143 creates an upper reflected electron image, a lower reflected electron image, a left reflected electron image, and a right reflected electron image. On the other hand, when creating a second X-direction magnetic domain contrast enhanced image, an upper reflected electron image, a left reflected electron image, and a right reflected electron image are required, so the first reflected electron image creation unit 143 creates an upper reflected electron image, a left reflected electron image, and a right reflected electron image. In this case, the first reflected electron image creation unit 143 does not need to create a lower reflected electron image. In the following description, the first X-direction magnetic domain contrast enhanced image or the second X-direction magnetic domain contrast enhanced image is also referred to as an X-direction magnetic domain contrast enhanced image.

Before performing the above-described addition and subtraction of pixel values of the upper reflected electron image, the lower reflected electron image, the left reflected electron image, and the right reflected electron image, the second reflected electron image creation unit 144 may perform background processing to reduce (preferably remove) the background (unevenness in brightness) occurring in the upper reflected electron image, the lower reflected electron image, the left reflected electron image, and the right reflected electron image. As described above, in this embodiment, in the sample setting process of step S202, if the surface of the sample S is not parallel to the surface of the sample holder 112, a gradation may occur in a specific direction in the reflected electron image. In such a case, it is preferable that the second reflected electron image creation unit 144 performs background processing to reduce (preferably remove) the gradation. For example, the second reflected electron image creation unit 144 performs a process of calculating a pixel value (intensity) due to the background from the gradient of the gradation in the direction showing the gradation in the reflected electron image, and subtracting this from the pixel value of the original reflected electron image. The gradient of the gradation is expressed, for example, as the gradient of a linear function that linearly approximates the relationship between pixel position and pixel value due to the gradation in the direction indicating the gradation.

Returning to the explanation of FIG. 1 and FIG. 2, the magnetic domain structure acquisition unit 145 acquires information indicating the size, shape, and magnetization direction of each magnetic domain in the sample S using the contrast derived from the magnetic domains having magnetization directions corresponding to the [100] direction and the [-100] direction and the contrast derived from the magnetic domains having magnetization directions corresponding to the [010] direction and the [0-10] direction (magnetic domain structure acquisition process of step S207 in FIG. 2). Here, the information indicating the size, shape, and magnetization direction of each magnetic domain in the sample S is information of a region including an observation region of the crystal orientation of the sample, for example, information of the same region as the observation region of the crystal orientation of the sample. The information indicating the size, shape, and magnetization direction of each magnetic domain in the sample S may be, for example, an image with a different density or color for each magnetization direction. In the following explanation, this image is also referred to as a magnetic domain structure map. It is preferable that the magnetic domain structure map overlaps with the crystal orientation map 320. That is, it is preferable that corresponding pixels in the magnetic domain structure map and the crystal orientation map 320 indicate information about the same magnetic domain.

The magnetic domain structure acquisition unit 145 may display the X-direction magnetic domain contrast enhanced image and the Y-direction magnetic domain contrast enhanced image on a computer display. In this case, the analyst may refer to the X-direction magnetic domain contrast enhanced image and the Y-direction magnetic domain contrast enhanced image to identify the magnetic domain structure of the sample S (the size, shape, and magnetic domain direction of each magnetic domain) and create information indicating the size, shape, and magnetization direction of each magnetic domain in the sample S. In this case, the analyst may input the information indicating the size, shape, and magnetization direction of each magnetic domain in the sample S to the information processing device 140.

The magnetic domain structure acquisition unit 145 may also automatically create information indicating the size, shape, and magnetization direction of each magnetic domain in the sample S based on the X-direction magnetic domain contrast enhanced image and the Y-direction magnetic domain contrast enhanced image. For example, the magnetic domain structure acquisition unit 145 may identify the magnetic domain direction at each pixel based on the result of comparing the pixel value of each pixel of the X-direction magnetic domain contrast enhanced image with a threshold value and the result of comparing the pixel value of each pixel of the Y-direction magnetic domain contrast enhanced image with a threshold value, and create information indicating the size, shape, and magnetization direction of each magnetic domain in the sample S. The magnetic domain structure acquisition unit 145 may display the created information indicating the size, shape, and magnetization direction of each magnetic domain in the sample S on a computer display.

When creating a magnetic domain structure map, in addition to the X-direction magnetic domain contrast enhanced image and the Y-direction magnetic domain contrast enhanced image, other backscattered electron images may be considered. For example, at least one of the upper backscattered electron image, the lower backscattered electron image, the left backscattered electron image, and the right backscattered electron image may be considered. Also, a full-area backscattered electron image may be considered. The full-area backscattered electron image is a backscattered electron image created by storing, for each measurement point MP, the integrated values of all pixels of the EBSD image 310 obtained by irradiating the measurement point MP of the sample S with the electron beam probe EP in the area (pixel) 331 corresponding to the measurement point MP.

As described above, in this embodiment, a backscattered electron image of the sample S is created for at least a portion of the region above, below, to the left, and to the right of the center of the EBSD detector 120, and a backscattered electron image including at least one of the upper backscattered electron image and the lower backscattered electron image, and a backscattered electron image including the left backscattered electron image and the right backscattered electron image is created based on each backscattered electron image, and an X-direction magnetic domain contrast-enhanced image and a Y-direction magnetic domain contrast-enhanced image are created as backscattered electron images in which the magnetic domain contrast in the ±X direction or ±Y direction is enhanced and the unevenness contrast is relaxed compared to the backscattered electron image, and the magnetic domain structure of the sample S is obtained using the created X-direction magnetic domain contrast-enhanced image and the Y-direction magnetic domain contrast-enhanced image. Therefore, it is possible to image a magnetic domain having a specific magnetization direction by making it brighter (or darker), and further, it is possible to emphasize the magnetic domain contrast while relaxing the unevenness contrast by adding or subtracting each backscattered electron image. Therefore, highly accurate information on the magnetic domain structure (size and shape of the magnetic domain and magnetization direction within the magnetic domain) can be obtained without precisely flattening the surface of the sample S. It is also possible to obtain information on the crystal orientation from the analysis results of the Kikuchi band positions in the EBSD image. Therefore, the crystal orientation map and magnetic domains can be observed in the same area, and information on the crystal orientation of the steel sheet and information on the magnetic domain structure (size and shape of the magnetic domains, and magnetization direction within the magnetic domains) can be observed in the same field of view.

(Modification)
In this embodiment, the pixel values of the upper backscattered electron image are subtracted from the pixel values of the Y-direction magnetic domain contrast relaxed image to create a second X-direction magnetic domain contrast enhanced image in which the topographical contrast is relaxed and the magnetic domain contrast in the ±X direction is enhanced. However, as described above, the topographical contrast tends to be reversed between the upper backscattered electron image and the lower backscattered electron image. Furthermore, in the lower backscattered electron image, the brightest observed region corresponds to the magnetic domain having the magnetization direction in the +X direction (i.e., the direction corresponding to the [100] direction), and the darkest observed region corresponds to the magnetic domain having the magnetization direction in the −X direction (i.e., the direction corresponding to the [−100] direction). Therefore, the lower backscattered electron image contains information on the topographical contrast of the surface of the sample S in addition to the magnetic domains having the magnetization direction in the ±X direction. Therefore, by utilizing this, a third X-direction magnetic domain contrast-enhanced image in which the unevenness contrast is relaxed and the magnetic domain contrast in the ±X directions is enhanced may be created as one of the X-direction magnetic domain contrast-enhanced images by adding the pixel values of the lower backscattered electron image to the pixel values of the Y-direction magnetic domain contrast relaxed image. Therefore, the second backscattered electron image creation unit 144 may create a third X-direction magnetic domain contrast-enhanced image in addition to or in place of at least one of the first X-direction magnetic domain contrast-enhanced image and the second X-direction magnetic domain contrast-enhanced image. As described above, the unevenness contrast is reversed between the upper backscattered electron image and the lower backscattered electron image, so in the example shown in Fig. 9, the pixel values of the magnetic domains 901 and 902 in the lower backscattered electron image are 0 and 8 (=1+7), respectively. 9(a) and 9(b), by adding the pixel values (0, 8) of the magnetic domains 901 and 902 in the lower backscattered electron image to the pixel values (2, 14) of the magnetic domains 901 and 902 in the lower backscattered electron image, the magnetic domain contrast in the ±X directions in the third X-direction magnetic domain contrast enhanced image becomes 12 to 20 (=22-2), so that the magnetic domain contrast in the ±X directions can be further enhanced and the unevenness contrast can be reduced. In order to create such a third X-direction magnetic domain contrast enhanced image, a lower backscattered electron image, a left backscattered electron image, and a right backscattered electron image are necessary, so that the first backscattered electron image creation unit 143 creates a lower backscattered electron image, a left backscattered electron image, and a right backscattered electron image. In this case, the first backscattered electron image creation unit 143 does not need to create an upper backscattered electron image.

In addition, in this embodiment, the sample S is set in the sample holder 112 so that the [100] and [-100] directions of the sample S correspond to either the +X direction or the -X direction on the surface of the sample S, respectively, and the [010] and [0-10] directions of the sample S correspond to either the +Y direction or the -Y direction on the surface of the sample S, respectively, to create an X-direction magnetic domain contrast enhanced image (at least one of the first to third X-direction magnetic domain contrast enhanced images) and a Y-direction magnetic domain contrast enhanced image, and a magnetic domain structure map is created. However, as described above, the sample S may be set in the sample holder 112 so that the [100] and [-100] directions of the sample S correspond to either the +Y direction or the -Y direction on the surface of the sample S, respectively, and the [010] and [0-10] directions of the sample S correspond to either the +X direction or the -X direction on the surface of the sample S, respectively (i.e., the sample S is rotated 90° so that the ±X direction and the ±Y direction are interchanged), and an X-direction magnetic domain contrast enhanced image and a Y-direction magnetic domain contrast enhanced image may be created, and a magnetic domain structure map may be created. Also, an X-direction magnetic domain contrast enhanced image and a Y-direction magnetic domain contrast enhanced image may be created in each state in which the sample S is rotated 90° so that the ±X direction and the ±Y direction are interchanged in this way, and a magnetic domain structure map may be created using these X-direction magnetic domain contrast enhanced image and Y-direction magnetic domain contrast enhanced image. In this way, for example, information regarding the magnetic domain structure (size and shape of the magnetic domains, and magnetization direction within the magnetic domains) can be determined using an X-direction magnetic domain contrast-enhanced image and a Y-direction magnetic domain contrast-enhanced image created in each state after rotating the sample S by 90° so that the ±X direction and the ±Y direction are interchanged, thereby further improving the accuracy of the magnetic domain structure map (information regarding the magnetic domain structure). For example, whether the magnetic domain contrast obtained from the X-direction magnetic domain contrast enhanced image and the Y-direction magnetic domain contrast enhanced image created by setting the sample S in the sample holder 112 so that the [100] and [-100] directions of the sample S correspond to either the +X direction or the -X direction of the sample S surface, respectively, and the [010] and [0-10] directions of the sample S correspond to either the +Y direction or the -Y direction of the sample S surface, respectively, is appropriate can be confirmed using the X-direction magnetic domain contrast enhanced image and the Y-direction magnetic domain contrast enhanced image created by setting the sample S in the sample holder 112 so that the [100] and [-100] directions of the sample S correspond to either the +Y direction or the -Y direction of the sample S surface, respectively, and the [010] and [0-10] directions of the sample S correspond to either the +X direction or the -X direction of the sample S surface, respectively.

Next, examples will be described, but the present invention is not limited to the following examples.
A commercially available Fe-3 mass% Si alloy was used as a sample, and the crystal orientation was measured using an EBSD device in a scanning electron microscope after the surface was electrolytically polished. The sample was then cut out so that the surface was parallel to the (001) plane of the BCC structure, and the sample surface was finished by electrolytic polishing to remove surface processing distortion. Furthermore, in the EBSD measurement setup of Figure 1, the sample was attached to a sample holder (EBSD holder) so that the +X direction was parallel to the [100] direction and the +Y direction was parallel to the [010] direction.

The sample holder was inserted into a scanning electron microscope and tilted by 70°. The acceleration voltage of the electron gun of the scanning electron microscope was set to 25 kV. A focused electron beam probe was irradiated onto the sample, and an EBSD image was obtained for each measurement point by scanning. The EBSD images were saved as digital information together with the position information of the measurement points. The resolution of the EBSD images was 100 pixels x 100 pixels. The EBSD images were saved in 8 bits. Commercially available EBSD analysis software (OIM Data Collection 7, manufactured by EDAX) was used to calculate crystal orientation information (Euler angles) from the EBSD images.

Figure 10 shows an example of a crystal orientation map 1000 obtained by EBSD analysis. As mentioned above, the crystal orientation map 1000 is a color map of the orientation of crystal planes parallel to the sample surface. For convenience of notation, the map is grayscaled in Figure 10, but the crystal orientation map 1000 shown in Figure 10 is a single red color. Therefore, it can be seen that the crystal orientation observation area of the sample is a single crystal, and the surface is approximately parallel to the (001) plane.

As shown in FIG. 11, an EBSD image of 100 pixels x 100 pixels was divided into 5 equal parts vertically and horizontally on a computer to divide it into 25 divided regions, and the pixel values (pixel intensity) of the pixels in each region were integrated. In the EBSD image divided into 25 divided regions, three divided regions in the center at the top, bottom, left, and right ends were selected, and the pixel intensities in these three regions were calculated. These were then arranged for each measurement point to form a backscattered electron image. FIG. 11(a) shows the divided regions selected when creating an upper backscattered electron image, FIG. 11(b) shows the divided regions selected when creating a lower backscattered electron image, FIG. 11(c) shows the divided regions selected when creating a left backscattered electron image, and FIG. 11(d) shows the divided regions selected when creating a right backscattered electron image. In FIG. 11(a) to FIG. 11(d), the three divided regions shown in darker shades indicate the selected divided regions. A divided area was also set in the EBSD image to correspond to the divided area of the EBSD detector, and the pixel values of the three divided areas of the EBSD image corresponding to the three divided areas shown dark in Figures 11(a) to 11(d) were integrated and stored as the pixel value of the measurement point for each measurement point of the sample, thereby creating an upper reflected electron image, a lower reflected electron image, a left reflected electron image, and a right reflected electron image. Background processing was performed on the upper reflected electron image, the lower reflected electron image, the left reflected electron image, and the right reflected electron image to remove diagonal gradation. In the following description, it is assumed that the upper reflected electron image, the lower reflected electron image, the left reflected electron image, and the right reflected electron image have been subjected to background processing.

Figure 12 shows an upper backscattered electron image 1200 formed only from backscattered electrons that reached above the center of the EBSD detector (the three divided regions in the center at the upper end shown in Figure 11(a)). In the upper backscattered electron image 1200 shown in Figure 12, a topographical contrast resulting from the topographical irregularities on the sample surface was obtained, and the magnetic domain structure was not clearly visible. This is thought to be due to the reasons mentioned above (Reason 1 to Reason 2).

Figure 13 shows a lower backscattered electron image 1300 formed only by the backscattered electrons that reached below the center of the EBSD detector (the three divided regions in the center at the lower end shown in Figure 11 (b)). In the lower backscattered electron image 1300 shown in Figure 13, a dendritic magnetic domain structure is clearly observed, with the brightly observed regions corresponding to magnetic domains whose magnetization direction is in the -X direction and the darkly observed regions corresponding to magnetic domains whose magnetization direction is in the +X direction. However, since a concavo-convex contrast is superimposed in addition to the magnetic domain contrast, it cannot be said that the contrast of the backscattered electron image is formed only by the magnetic domain contrast.

Therefore, taking into consideration the unevenness contrast obtained in the upper reflected electron image 1200 shown in FIG. 12, the unevenness contrast was removed as much as possible from the lower reflected electron image 1300 shown in FIG. 13, resulting in the first X-direction magnetic domain contrast-enhanced image 1400 shown in FIG. 14. The first X-direction magnetic domain contrast-enhanced image 1400 is obtained by adding the pixel values of the upper reflected electron image 1200 shown in FIG. 12 and the pixel values of the lower reflected electron image 1300 shown in FIG. 13 in order to reduce the unevenness contrast. In the first X-direction magnetic domain contrast-enhanced image 1400 shown in FIG. 14, the unevenness contrast is reduced and the magnetic domain contrast in the ±X directions is clearer than in the lower reflected electron image 1300 shown in FIG. 13. On the other hand, a reflected electron image created by subtracting the pixel values of the lower reflected electron image 1300 shown in FIG. 13 from the pixel values of the upper reflected electron image 1200 shown in FIG. 12 is shown in FIG. 15. In the backscattered electron image shown in FIG. 15, the unevenness contrast is not reduced and the magnetic domain contrast in the ±X directions is not emphasized, compared to the first X-direction magnetic domain contrast-enhanced image 1400 shown in FIG. 14.

Next, FIG. 16 shows a left-side backscattered electron image 1600 formed only by backscattered electrons that reach the left side of the center of the EBSD detector (the three divided areas in the center at the left end shown in FIG. 11(c)). FIG. 17 shows a right-side backscattered electron image 1700 formed only by backscattered electrons that reach the right side of the center of the EBSD detector (the three divided areas in the center at the right end shown in FIG. 11(d)).

In the left backscattered electron image 1600 and the right backscattered electron image 1700 shown in Figs. 16 and 17, the dendritic magnetic domain structure is clearly observed, but it can be seen that the contrast is different from that of the lower backscattered electron image 1300 shown in Fig. 13. This is because, as described above, the magnetic domains that appear bright can be changed depending on which part of the EBSD detector the electrons reaching are used to construct the backscattered electron image. The areas in which the contrast appears bright in the left backscattered electron image 1600 and the right backscattered electron image 1700 shown in Figs. 16 and 17 is reversed correspond to areas (= magnetic domains) in which the magnetization direction is reversed in the ±Y direction. Fig. 18 shows a Y-direction magnetic domain contrast-enhanced image 1800 created by subtracting the pixel values of the right backscattered electron image 1700 shown in Fig. 17 from the pixel values of the left backscattered electron image 1600 shown in Fig. 16. In the Y-direction magnetic domain contrast-enhanced image 1800 shown in FIG. 18, the unevenness contrast is reduced and the magnetic domain contrast is enhanced compared to the left-side backscattered electron image 1600 shown in FIG. 16 and the right-side backscattered electron image 1700 shown in FIG. 17. In the Y-direction magnetic domain contrast-enhanced image 1800, the darkest magnetic domains are those magnetized in the +Y direction, and the brightest regions are those magnetized in the -Y direction.

Next, FIG. 19 shows a Y-direction magnetic domain contrast relaxed image 1900 created by adding the pixel values of the left-side reflected electron image 1600 shown in FIG. 16 and the right-side reflected electron image 1700 shown in FIG. 17. In the Y-direction magnetic domain contrast relaxed image 1900 shown in FIG. 19, the magnetic domain contrast (magnetic domains whose magnetization directions are +Y and -Y directions) that are inverted in the left-side reflected electron image 1600 shown in FIG. 16 and the right-side reflected electron image 1700 shown in FIG. 17 are added together, so that the magnetic domain contrast is relaxed, and the magnetic domains whose magnetization directions are +X and -X directions can be emphasized. However, in the Y-direction magnetic domain contrast relaxed image 1900, the magnetic domain contrast in the ±X directions is difficult to observe because the concavo-convex contrast is superimposed on the magnetic domain contrast in the ±X directions. Therefore, by subtracting the pixel values of the upper reflected electron image 1200 shown in FIG. 12 from the pixel values of the Y-direction magnetic domain contrast relaxed image 1900 shown in FIG. 19, a second X-direction magnetic domain contrast enhanced image 2000 shown in FIG. 20 is obtained, in which the unevenness contrast is relaxed and the magnetic domain contrast in the ±X directions is further enhanced.

From the above results, the magnetic domain structure map 2100 shown in Figure 21 is a diagram that visualizes the area of each magnetic domain within the observation area of the magnetic domain structure of the sample and the magnetization direction within each magnetic domain. The observation area of the crystal orientation of the sample and the observation area of the magnetic domain structure are the same. Therefore, the crystal orientation map 1000 shown in Figure 10 and the magnetic domain structure map 2100 shown in Figure 21 are of exactly the same area. Therefore, the crystal orientation and the magnetic domain structure can be observed in the same field of view.

Figure 22 shows the result of forming a backscattered electron image (normal backscattered electron image) using only the backscattered electrons that reached the entire EBSD detector. In the full-area backscattered electron image 2200 shown in Figure 22, the topography contrast is superimposed on the magnetic domain contrast, and this image alone does not provide any insight into the magnetization direction within the magnetic domains.

The above-described embodiments and examples of the present invention are merely examples of how the present invention can be implemented, and the technical scope of the present invention should not be interpreted in a limiting manner. In other words, the present invention can be implemented in various forms without departing from its technical concept or main features.

110 Scanning electron microscope 111 Optical system 112 Sample holder 120 EBSD detector 130 Imaging device 140 Information processing device 141 Image acquisition section 142 Crystal orientation acquisition section 143 First reflected electron image creation section 144 Second reflected electron image creation section 145 Magnetic domain structure acquisition section 310 EBSD image 320 Crystal orientation map 330 Reflected electron image 510 Arrival position of reflected electrons when no internal magnetic field exists in the sample 511 Arrival position of reflected electrons when an internal magnetic field in the +X direction exists in the sample 512 Arrival position of reflected electrons when an internal magnetic field in the -X direction exists in the sample 611 Arrival position of reflected electrons when an internal magnetic field in the +Y direction exists in the sample 612 Arrival position of reflected electrons when an internal magnetic field in the -Y direction exists in the sample 701-702 Magnetic domains whose magnetization directions are adjacent to each other in the ±X direction 801-802 Magnetic domains adjacent to each other with magnetization directions in ±Y directions 901-902 Magnetic domains adjacent to each other with magnetization directions in ±X directions 1000 Crystal orientation map 1200 Upper backscattered electron image 1300 Lower backscattered electron image 1400 First X-direction magnetic domain contrast-enhanced image 1600 Left-side backscattered electron image 1700 Right-side backscattered electron image 1800 Y-direction magnetic domain contrast-enhanced image 1900 Y-direction magnetic domain contrast relaxation image 2000 Second X-direction magnetic domain contrast-enhanced image 2100 Magnetic domain structure map 2200 Full-area backscattered electron image BE Backscattered electron DU Travel distance within the sample of the backscattered electron heading upward DD Travel distance within the sample of the backscattered electron heading downward EP Electron beam probe S Sample

Claims (7)

A method for observing magnetic domains in a steel sheet, the method comprising: irradiating a measurement point of a sample made of a steel sheet having a body centered cubic (BCC) structure with an electron beam in an electron microscope; and using an electron backscatter diffraction (EBSD) image showing a diffraction pattern of reflected electrons that reach an EBSD detector, the method comprising the steps of:
In the EBSD detector, when the electron beam is projected onto the EBSD detector, a direction parallel to the irradiation direction of the projected electron beam is defined as an irradiation parallel direction, a direction anti-parallel to the irradiation parallel direction is defined as an irradiation anti-parallel direction, and two directions perpendicular to the irradiation parallel direction and anti-parallel to each other are defined as a first irradiation side direction and a second irradiation side direction, respectively;
a sample preparation step of preparing the sample so that the sample surface corresponds to the (001) plane of the BCC structure;
a sample setting step of setting the sample in the electron microscope such that the [100] direction and the [−100] direction of the crystal correspond to one of the irradiation parallel direction and the irradiation anti-parallel direction, and the first irradiation side direction and the second irradiation side direction, and the [010] direction and the [0−10] direction of the crystal correspond to the other;
a first backscattered electron image creating step of calculating an intensity of backscattered electrons in a partial region of the EBSD image for the sample set in the sample setting step while varying the measurement point, thereby creating a backscattered electron image in the partial region of the EBSD image;
a second backscattered electron image creating step of creating a backscattered electron image based on the backscattered electron image created in the first backscattered electron image creating step, in which a contrast derived from magnetic domains having magnetization directions corresponding to the [100] direction and the [−100] direction or a contrast derived from magnetic domains having magnetization directions corresponding to the [010] direction and the [0-10] direction is emphasized and a contrast derived from the shape of the sample surface is alleviated compared to the backscattered electron image created in the first backscattered electron image creating step;
a magnetic domain structure acquiring step of acquiring information indicating the size, shape, and magnetization direction of each magnetic domain in the sample using the backscattered electron image created in the second backscattered electron image creating step;
having
The first reflected electron image creation step creates at least one of an anti-parallel-irradiation direction side reflected electron image which is a reflected electron image in at least a partial region of the EBSD image on the anti-parallel-irradiation direction side of the position corresponding to the measurement point and a parallel-irradiation direction side reflected electron image which is a reflected electron image in at least a partial region of the EBSD image on the parallel-irradiation direction side of the position corresponding to the measurement point, a first irradiation lateral direction side reflected electron image which is a reflected electron image in at least a partial region of the EBSD image on the first irradiation lateral direction side of the position corresponding to the measurement point, and a second irradiation lateral direction side reflected electron image which is a reflected electron image in at least a partial region of the EBSD image on the second irradiation lateral direction side of the position corresponding to the measurement point. The method for observing magnetic domains in steel sheet according to claim 1, wherein the first backscattered electron image creation process creates a backscattered electron image in the anti-parallel direction of the irradiation, a backscattered electron image in the parallel direction of the irradiation, a backscattered electron image in the first side direction of the irradiation, and a backscattered electron image in the second side direction of the irradiation. The method for observing magnetic domains in a steel sheet according to claim 2, wherein, in the second backscattered electron image creation step, by performing addition of the pixel values of the backscattered electron image in the anti-parallel direction to the pixel values of the backscattered electron image in the parallel direction, a first-irradiation-direction magnetic domain contrast-enhanced image is created, which is a backscattered electron image in which the contrast originating from the magnetic domains having a magnetization direction corresponding to one of the [100] and [-100] directions, and the [010] and [0-10] directions is enhanced and the contrast originating from the shape of the sample surface is relaxed, and by performing subtraction of the pixel values of the backscattered electron image in the first-irradiation-side direction from the pixel values of the backscattered electron image in the second-irradiation-side direction, a backscattered electron image in which the contrast originating from the magnetic domains having a magnetization direction corresponding to the other direction is enhanced and the contrast originating from the shape of the sample surface is relaxed. 2. The method for observing magnetic domains in a steel sheet according to claim 1 , wherein in the first reflected electron image creation process, a reflected electron image in the anti-parallel direction to the irradiation, a reflected electron image in the first lateral direction to the irradiation, and a reflected electron image in the second lateral direction to the irradiation are created, and the reflected electron image in the parallel direction to the irradiation is not created. In the second backscattered electron image creation step, the pixel values of the backscattered electron image in the first side direction of irradiation and the pixel values of the backscattered electron image in the second side direction of irradiation are added together, and then the pixel values of the backscattered electron image in the anti-parallel direction of irradiation are subtracted to create a second backscattered electron image in the direction of irradiation, in which the contrast originating from the magnetic domains having a magnetization direction corresponding to one of the [100] and [-100] directions, and the contrast originating from the shape of the sample surface is relaxed, and a backscattered electron image in the direction of irradiation, in which the contrast originating from the magnetic domains having a magnetization direction corresponding to the other direction is emphasized and the contrast originating from the shape of the sample surface is relaxed, is created. The magnetic domain observation method in a steel sheet according to claim 2 or 4. The method for observing magnetic domains in steel sheet according to any one of claims 1 to 5, wherein in the first backscattered electron image creation step, the EBSD image is divided into a plurality of regions such that at least three regions exist in each of the directions corresponding to the parallel irradiation direction and the anti-parallel irradiation direction, and the directions corresponding to the first irradiation side direction and the second irradiation side direction, and a backscattered electron image is created in each of the partial regions of the EBSD image by using pixel values of the divided regions located closest to the edge of the EBSD image, without using pixel values of the divided regions located closest to the position corresponding to the measurement point in the EBSD image. the backscattered electron image created in the first backscattered electron image creating step is an image including an observation region of a crystal orientation of the sample that is analyzed based on the EBSD image,
The method for observing magnetic domains in a steel sheet according to any one of claims 1 to 6, wherein the information indicating the size, shape, and magnetization direction of each magnetic domain in the sample is information of a region including an observation region of the crystal orientation of the sample.

Images