Three-dimensional observation technique, processing thickness calculation method, and sample

Abstract

To provide a three-dimensional observation technique, a processing thickness calculation method, and a sample that can improve the accuracy of specifying coordinate positions and resolution in a depth direction of a plurality of cross-sectional images.SOLUTION: A three-dimensional observation technique includes: an observation step of acquiring a plurality of cross-sectional images orthogonal to a depth direction (Z direction) that is one direction of a sample 2; a correction step of providing the coordinates in the depth direction to the plurality of cross-sectional images; and a structure acquisition step of acquiring the three-dimensional structure of the sample from the plurality of cross-sectional images. The cross-sectional images are charged particle images or optical images.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a three-dimensional observation method, a method for calculating a processing thickness, and a sample.

Background Art

[0002] For example, when acquiring the internal structure of a sample having a three-dimensional structure such as a porous body, an observation method called serial sectioning is known. In this method, a cross-section of the sample is formed by processing such as polishing or cutting, and an observation image of this cross-section is acquired. Further, by repeating this processing and acquisition of the observation image, a continuous cross-sectional image of the sample is acquired. Then, by assigning coordinates and correcting each cross-sectional image in a computer, cross-sectional information at an arbitrary position of the sample can be obtained.

[0003] In the processing of serial sectioning, in addition to mechanical polishing and cutting, processing using a charged particle beam device can be used. (For example, Patent Document 1, Non-Patent Document 1). Further, as the observation method, a charged particle image or an optical image can be used depending on the required information.

[0004] According to such an observation method, by three-dimensionally constructing the cross-sectional image stack obtained in the above-described process in a computer, an arbitrary cross-section other than the observed cross-section can be constructed. Therefore, in addition to obtaining a three-dimensional image of the sample, prediction of physical properties by computer simulation can also be performed.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Non-Patent Documents

[0006]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0007] By the way, in the conventional serial sectioning method, a new sample cross-section is formed by cutting the sample to a predetermined thickness, but there is a risk that the processing thickness may not be as set due to the apparatus conditions. And when the processing thickness does not become as set, there is a problem that the accuracy and resolution of the coordinate position of the image in the depth direction, which is the direction in which the processed surfaces (cross-sections) of the sample are arranged, decrease.

[0008] Therefore, the present invention has been made paying attention to the above problems, and an object thereof is to provide a three-dimensional observation method, a method for calculating the processing thickness, and a sample capable of improving the specific accuracy and resolution of the coordinate position in the depth direction of a plurality of cross-sectional images.

Means for Solving the Problems

[0009] (1) According to one aspect of the present invention, there is provided a three-dimensional observation method including an observation step of acquiring a plurality of cross-sectional images orthogonal to the depth direction, which is one direction of the sample, a correction step of giving coordinates in the depth direction to the plurality of cross-sectional images, and a structure acquisition step of acquiring a three-dimensional structure of the sample from the plurality of cross-sectional images using the coordinates in the depth direction, wherein the cross-sectional images are charged particle images or optical images.

[0010] (2) In the three-dimensional observation method according to (1) above, a pattern in which at least one of the shape and the formation position changes as the sample changes in the depth direction is formed on the surface or inside of the sample, and in the correction step, coordinates in the depth direction are given to the plurality of cross-sectional images based on the change of the pattern.

[0011] (3) In the three-dimensional observation method according to (2) above, a reference pattern extending in the depth direction is formed on the surface or inside of the sample.

[0012] (4) In the three-dimensional observation method according to (2) or (3) above, before the observation step, a pattern forming step of forming the pattern on the sample is further provided.

[0013] (5) In the three-dimensional observation method according to any one of (1) to (4) above, in the observation step, the sample is cut in the depth direction, and the charged particle image or the optical image of the cut cross section of the sample is repeatedly obtained to obtain the plurality of cross-sectional images.

[0014] (6) According to one aspect of the present invention, an observation step of obtaining a plurality of cross-sectional images orthogonal to the depth direction by repeatedly cutting a sample in the depth direction, which is a single direction, and obtaining a charged particle image or an optical image of the cut cross section of the sample; a correction step of giving coordinates in the depth direction to the plurality of cross-sectional images; and a method for calculating a machining thickness for calculating a machining thickness when cutting the sample using the coordinates in the depth direction are provided.

[0015] (7) According to one aspect of the present invention, a sample having a three-dimensional structure, and having a pattern on the surface or inside, at least one of the shape and the formation position of which changes in accordance with a change in a single direction, the depth direction, is provided.

Effects of the Invention

[0016] According to one aspect of the present invention, a three-dimensional observation method, a method for calculating a machining thickness, and a sample are provided, which can improve the specific accuracy and resolution of the coordinate positions in the depth direction of a plurality of cross-sectional images.

Brief Description of the Drawings

[0017]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Mode for Carrying Out the Invention

[0018] In the following detailed description, embodiments of the present invention will be described with reference to the drawings. In the description of the drawings, the same or similar parts are denoted by the same or similar reference numerals, and duplicate descriptions are omitted. Each drawing is schematic and may include cases that are different from the actual ones. Further, the embodiments shown below are examples of devices and methods for embodying the technical idea of the present invention, and the technical idea of the present invention does not specify the materials, structures, corrections, etc. of the components as the following. The technical idea of the present invention can be variously modified within the technical scope defined by the claims described in the claims.

[0019] A three-dimensional observation method according to an embodiment of the present invention will be described. In this embodiment, serial sectioning using a FIB-SEM composite machine is performed as the three-dimensional observation method. As shown in FIG. 1, the FIB-SEM composite machine 1 is an apparatus for three-dimensionally observing a sample 2 having a three-dimensional structure, and includes a scanning electron microscope (SEM) 10, a focused ion beam (FIB) processing apparatus 11, and a secondary electron detector 12. The FIB-SEM composite machine 1 cuts the surface of the sample 2 to a predetermined depth with the focused ion beam processing apparatus 11, and acquires an image of the cut cross section (cross-sectional image) as a charged particle image with the scanning electron microscope 10. Then, by repeating this cutting and image acquisition, a plurality of cross-sectional images continuous in the depth direction of the sample 2 are acquired. In this embodiment, the depth direction is one direction in the sample 2 and is the cutting direction in which the cutting of the sample 2 is continuously performed. That is, the depth direction is a direction orthogonal to the plurality of cross-sectional images. Also, in this embodiment, the depth direction is also referred to as the Z-axis direction. Furthermore, the two directions that are orthogonal to the Z-axis direction and orthogonal to each other are also referred to as the X-axis direction and the Y-axis direction. The X-axis direction and the Y-axis direction are directions parallel to the cross-sectional image.

[0020] In the three-dimensional observation method according to this embodiment, as shown in FIG. 2, first, preprocessing of the sample 2 is performed (S100, preprocessing step). In the preprocessing, excess portions of the sample 2 are removed, and a layer such as carbon is provided on the surface of the sample 2 for protection of the sample 2.

[0021] Next, a pattern is formed on the surface of the sample 2 (S102, pattern formation step). The pattern is such that at least one of the shape and the formation position changes as the depth direction changes, and is provided in the region where three-dimensional observation is to be performed. Also, the pattern may be formed by attaching a mesh (lattice-shaped marking member) having a regular pattern, or by the focused ion beam processing apparatus 11.

[0022] In this embodiment, as an example, a pattern 3 as shown in FIG. 3 is formed. In the example shown in FIG. 3, the pattern 3 is composed of a plurality of linear grooves inclined at a predetermined angle with respect to the Z-axis direction. In FIG. 3, the region to be cut for obtaining the cross-sectional image is a rectangular region D indicated by a broken line. The angle θ (tilt angle) of the pattern 3 with respect to the Z-axis direction is not particularly limited, but it is preferably 45 degrees or more and 80 degrees or less on the surface of the sample 2 in the Z-axis direction. More preferably, the angle θ is 54 degrees or more and 75 degrees or less. When the angle θ is less than 45 degrees, the change of the pattern 3 in the depth direction becomes small, so the setting accuracy of the coordinates in the depth direction may decrease depending on the cutting interval. On the other hand, when the angle θ exceeds 80 degrees, the width of the groove appearing in the cross section becomes large, so the setting accuracy of the coordinates in the depth direction described later may decrease. Since the change of the pattern 3 in the depth direction becomes small, the setting accuracy of the coordinates in the depth direction may decrease depending on the cutting interval. Therefore, by setting the angle θ to 45 degrees or more and 80 degrees or less, the setting accuracy of the coordinates in the depth direction can be further improved, and by setting it to 54 degrees or more and 75 degrees or less, further accuracy improvement can be expected. Also, the number of grooves is not particularly limited, but it is preferably set so that a plurality of grooves appear in the cross-sectional image.

[0023] Also, in step S102, a reference pattern 30, which is a linear groove extending in the depth direction, is formed on the surface of the sample 2 in the same manner as the pattern 3. The reference pattern 30 may be at least one, but in order to improve the adjustment accuracy of the image shift in the X-axis direction and the Y-axis direction described later, it is preferably two or more. In the example shown in FIG. 3, two reference patterns 30 are formed sandwiching the pattern 3 in the X-axis direction.

[0024] Furthermore, after step S102, the FIB cutting conditions for serial sectioning are set, and on top of that, the SEM observation conditions are set (S104, condition setting step). The FIB cutting conditions are set, for example, with a processing thickness of 10 nm to 300 nm.

[0025] Thereafter, serial sectioning is performed by repeatedly cutting sample 2 in the depth direction and acquiring charged particle images of the cut cross-section of sample 2 (S106, observation step). FIG. 4 shows an example of a cross-sectional image obtained in the observation step. As shown in FIG. 4, a cross-sectional image is obtained in which a plurality of grooves formed by pattern 3 are formed between the grooves of reference pattern 30. In the present embodiment, as will be described later, three-dimensional observation with excellent resolution in the depth direction can be performed. For this reason, in step S106, by setting the cutting thickness (processing thickness) during cutting to 5 nm or less, measurement with an accuracy exceeding the resolution of the observation image becomes possible.

[0026] Next, correction of the image position in the XY direction is performed on the plurality of cross-sectional images acquired in step S106 (S108, XY correction step). This correction adjusts the deviation in the X-axis direction and Y-axis direction of each cross-sectional image. For example, in step S108, correction may be performed to rotate and move each cross-sectional image in the X-Y plane so that the grooves of reference pattern 30 are aligned in the plurality of cross-sectional images. In step S106, it is preferable that an image obtained by performing binarization processing on the cross-sectional image is used as the cross-sectional image.

[0027] Furthermore, a process of extracting the structure of pattern 3 is performed on the plurality of cross-sectional images corrected in the XY direction (S110, pattern extraction step). In step S110, correction in the XY direction is performed, and for the plurality of binarized cross-sectional images, the X coordinates and Y coordinates, which are the coordinates in the X-axis direction and Y-axis direction of the plurality of pattern 3s in each cross-sectional image, are obtained.

[0028] Thereafter, based on the obtained X coordinates and Y coordinates, the movement amount of pattern 3 is measured (S112, movement amount calculation step). Step S112 is performed, for example, by the image phase difference method, and the movement amount of pattern 3, that is, the movement amount in the X-axis direction of the groove of pattern 3 in the continuous front and rear cross-sectional images, is measured from the position of the groove of the same pattern 3 in the continuous cross-sectional images.

[0029] Next, based on the measured movement amount of Pattern 3, the image position in the Z-axis direction is corrected, thereby giving the cross-sectional image the coordinate (Z coordinate) in the depth direction (S114, correction step). Pattern 3 moves in the X-axis direction as the Z-axis changes. Therefore, from the movement amount of Pattern 3, the relative position in the Z-axis direction between consecutive cross-sectional images can be known, and the position of each cross-sectional image in the Z-axis direction can be accurately corrected. Also, in step S114, the cutting thickness (processing thickness) is calculated from the Z coordinates of a plurality of cross-sectional images corrected in the Z-axis direction.

[0030] Furthermore, by reconstructing a plurality of cross-sectional images whose coordinates in the Z-axis direction have been corrected, the three-dimensional structure of the sample 2 is obtained (S116, structure acquisition step).

[0031] As described above, according to the present embodiment, it includes an observation step of acquiring a plurality of cross-sectional images orthogonal to the depth direction, which is one direction of the sample 2, a correction step of giving the coordinates in the depth direction to the plurality of cross-sectional images, and a structure acquisition step of acquiring the three-dimensional structure of the sample 2 from the plurality of cross-sectional images using the coordinates in the depth direction. With such a configuration, since the coordinates in the depth direction are corrected to obtain the three-dimensional structure, even when there is variation in the processing thickness, a highly accurate three-dimensional structure can be obtained.

[0032] Also, during continuous cross-sectional observation by serial sectioning, the cutting thickness measurement step (real-time measurement step) for each cutting process can be omitted, thereby improving the throughput during observation. For example, when the real-time measurement step is omitted as described above and the data without measuring and correcting the cutting thickness is used as it is, the three-dimensional structure reconstructed from a large number of cross-sectional images will include distortion due to the instability of the device accuracy. However, in the subsequent process after the observation step, by measuring the cutting thickness and correcting the coordinate position, a three-dimensional structure without distortion can be obtained.

[0033] Furthermore, according to the present embodiment, since the cutting thickness is measured using the pattern movement amount that does not depend on the resolution of the observation image, measurement with an accuracy exceeding the resolution of the observation image can be performed. For example, in a surface SIM image generally acquired for measuring the cutting position, the resolution is only about 10 nm, so the measurement accuracy and the position accuracy of the beam at the time of the next cutting can only be improved to about 10 nm. Although the resolution of FE-SEM as used in the present embodiment also remains at about several nm, by measuring the movement amount of the pattern in the FE-SEM image, the cutting thickness can be determined with an accuracy of 5 nm or less, and further 1 nm or less.

[0034] Furthermore, according to the present embodiment, since a SIM image is not acquired after cutting, in addition to suppressing damage to the surface of the sample 2, a significant variation in the scan coil current of the FIB can be suppressed. For this reason, an improvement in the position stability of the ion beam can be expected. While the feed unit in the cutting direction (Z-axis) of serial sectioning by general FIB-SEM is only about 10 nm, by using the present invention, it can be increased to 5 nm or less.

[0035] The three-dimensional observation method according to the present embodiment is particularly suitable for observing a sample 2 having a fine three-dimensional structure because the resolution is on the order of nanometers. The three-dimensional observation method according to the present embodiment is suitable for observing samples that require dimensional accuracy, such as porous bodies, LSIs, and automotive battery members, for example.

[0036] <Modification Example> Although the present invention has been described with reference to specific embodiments above, it is not intended to limit the invention by these descriptions. By referring to the description of the present invention, those skilled in the art will also understand other embodiments of the present invention including various modification examples together with the disclosed embodiments. Therefore, the embodiments of the invention described in the claims should be construed to cover embodiments including these modification examples described herein alone or in combination.

[0037] For example, in the above embodiment, Pattern 3 is shown in FIG. 3, but the present invention is not limited to such an example. Pattern 3 may be anything that moves in the X-axis direction as the Z-axis direction changes. For example, as shown in FIG. 5(A), Pattern 3 may be a V-shaped groove in which the distance between two straight lines increases as it moves in the depth direction. Also, as shown in FIG. 5(B), Pattern 3 may be a diamond-shaped hole. Furthermore, as shown in FIG. 5(C), Pattern 3 may be a plurality of dot-shaped holes.

[0038] Also, in the above embodiment, it is assumed that the reference pattern 30 is used, but the present invention is not limited to such an example. If correction in the X-axis direction and the Y-axis direction is possible even without the reference pattern 30, the reference pattern 30 may not be formed. For example, in the case of the V-shaped Pattern 3 shown in FIG. 5(A), correction in the X-axis direction and the Y-axis direction can be performed by aligning the centers of the two grooves formed in the cross-section. Also, if there is no deviation in the X-axis direction and the Y-axis direction from the mechanism at the time of image acquisition, correction in the X-axis direction and the Y-axis direction may not be performed.

[0039] Furthermore, in the above embodiment, the case of serial sectioning using a FIB-SEM composite machine has been described, but the present invention is not limited to such an example. The present invention can also be applied to other aspects as long as it is a method for acquiring a three-dimensional structure that repeatedly removes a sample by cutting or the like at a predetermined thickness and observes the removed cross-section in one direction. It can also be applied in the case of general serial sectioning methods other than the method using a FIB-SEM composite machine, for example, when creating a cross-section by etching the metal surface and obtaining an optical image using an optical microscope.

[0040] Furthermore, in the above embodiment, it is assumed that Pattern 3 is formed by cutting the surface of Sample 2, but the present invention is not limited to such an example. Pattern 3 may be formed on the surface or inside of Sample 2. For example, a modified example of the case where a film 4 such as Pt is formed on the surface of Sample 2 is shown for Pattern 3. Among these, Fig. 6(A) shows the case where a film 4 is formed on the surface of Sample 2 and Pattern 3 is formed on the surface of this film 4. For example, when the surface of Sample 2 is rough or the like, it may be difficult to form Pattern 3 on the surface of Sample 2. However, even in such a case, as shown in Fig. 6(A), by forming a film 4 on the surface of Sample 2 and forming Pattern 3 on this film 4, three-dimensional observation can be performed with high accuracy. Also, as shown in Fig. 6(B), an embodiment in which a film 4 is formed on the surface of Sample 2 after forming Pattern 3 on the surface of Sample 2 may be used.

[0041] Furthermore, the present invention can also be applied to Sample 2 having a three-dimensional structure. Such a Sample 2 has a Pattern 3 in which at least one of the shape and the formation position changes along with a change in the depth direction, which is a one-way direction, on the surface or inside. Also, Sample 2 can be configured as described in the above embodiment and other modified examples.

Example

[0042] Examples conducted by the present inventors will be described. In the examples, serial sectioning was performed using a focused ion beam (FIB) processing apparatus equipped with a scanning electron microscope (SEM). At this time, as in the above embodiment, Pattern 3 was formed on the surface of Sample 2 using the FIB processing apparatus. An electron microscope image of Sample 2 on which Pattern 3 was formed is shown in Fig. 7. The inclination angle θ of Pattern 3 with respect to the Z-axis direction was set to 72 degrees, and a pattern was formed in the sample such that cross-sections of five inclined grooves could always be obtained in the cross-sectional image in the Z-axis direction. The material of the sample is a Si wafer.

[0043] Then, serial sectioning was performed in the same manner as in the above embodiment to obtain a plurality of cross-sectional images. In this example, the cutting thickness was set to 5 nm and serial sectioning was performed. FIG. 8 shows an electron microscope image during serial sectioning performed in the manner of FIG. 6(B). As shown in FIG. 8, it was confirmed that the pattern 3 shown in FIG. 7 was included in the cross section. And it was also confirmed that the region where three-dimensional observation was desired to be performed could be included below the pattern 3.

[0044] FIG. 9 shows an image reconstructed in a computer such that the positions of the Z-axis coordinates are equally spaced in the order of acquisition of the obtained plurality of cross-sectional images. As shown in FIG. 9, it was observed that the pattern 3, which was linear as shown in FIG. 7, became a distorted pattern. Also, it was observed that the pitch intervals between the grooves also became uneven intervals.

[0045] FIG. 10 shows the result of measuring the positions of the grooves of the pattern 3 from the obtained plurality of cross-sectional images and calculating the cutting thickness from the amount of movement of the grooves between the images. As shown in FIG. 10, although the set cutting thickness was 5 nm, as a result of the measurement, it was confirmed that the cutting thickness fluctuated around 5 nm. It can be seen that such a variation in the cutting thickness results in the distortion in FIG. 9.

[0046] FIG. 11 shows the result of reflecting the numerical values (Z-axis coordinates) of the cutting thickness in each of the cross-sectional images shown in FIG. 10 in a three-dimensional space in a computer, that is, the result of correcting the coordinates in the Z-axis direction in the same manner as in the above embodiment. As shown in FIG. 11, according to the above embodiment, it was confirmed that, similar to FIG. 7, the pattern 3 formed on the sample surface was accurately represented, and the measurement of the cutting thickness and the correction of the Z-axis coordinates could be accurately performed. Therefore, it was confirmed that an accurate three-dimensional structure could also be obtained for the sample volume for three-dimensional observation located below the pattern 3.

Explanation of Signs

[0047] 1 FIB-SEM composite machine 10 Scanning electron microscope (SEM) 11 Focused ion beam (FIB) processing apparatus 12 Secondary electron detector 2 Specimen 3 Pattern 30 Reference pattern

Claims

1. An observation step of acquiring a plurality of cross-sectional images orthogonal to the depth direction which is one direction of the sample, A correction step of giving coordinates in the depth direction to the plurality of cross-sectional images, A structure acquisition step of acquiring a three-dimensional structure of the sample from the plurality of cross-sectional images using the coordinates in the depth direction, Comprising, A three-dimensional observation method in which the cross-sectional image is a charged particle image or an optical image.

2. On the surface or inside of the sample, a pattern is formed in which at least one of the shape and the formation position changes as the depth direction changes, The three-dimensional observation method according to claim 1, wherein in the correction step, coordinates in the depth direction are given to the plurality of cross-sectional images based on the change of the pattern.

3. The three-dimensional observation method according to claim 2, wherein a reference pattern extending in the depth direction is formed on the surface or inside of the sample.

4. The three-dimensional observation method according to claim 2 or 3, further comprising a pattern formation step of forming the pattern on the sample before the observation step.

5. In the observation step, the sample is cut in the depth direction, and the plurality of cross-sectional images are acquired by repeating the acquisition of the charged particle image or the optical image of the cut cross-section of the sample. The three-dimensional observation method according to any one of claims 1 to 3.

6. An observation step of cutting a sample in the depth direction which is one direction, and repeating the acquisition of a charged particle image or an optical image of the cut cross-section of the sample to acquire a plurality of cross-sectional images orthogonal to the depth direction, A correction step of giving coordinates in the depth direction to the plurality of cross-sectional images, A method for calculating the machining thickness, which calculates the machining thickness when cutting the sample using the coordinates in the depth direction.

7. A sample having a three-dimensional structure, On the surface or inside, there is a pattern in which at least one of the shape and the formation position changes as the depth direction which is one direction changes.

Images