6 of the integrated control unit C0 can compare the image data D1 and the image data D2 and calculate a shift amount indicating how much a position of the mesh MS in the image data D2 is shifted from a position of the mesh MS in the image data D1.

An example of a method for calculating the shift amount includes a method using the reference area MSa of the mesh MS. For example, with reference to FIGS. 8 and 9, how much a position of the reference area MSa in the image data D2 is shifted from a position of the reference area MSa in the image data D1 is calculated. That is, movement amounts (ΔX, ΔY) in the X direction and the Y direction in a plan view and a rotation angle (ΔΦ) in a plan view are measured. Values of ΔX, ΔY and ΔΦ are calculated as the shift amount.

The reference area MSa in the first embodiment is an area having a shape different from a shape of grids around the reference area MSa, and is formed near the center of the mesh MS. Alternatively, the reference area MSa may be any area as long as it can be identified that the reference area MSa corresponds to coordinates of which position of the mesh MS. For example, a specific mark formed on an outer periphery of the mesh MS, such as an area surrounding grids, may be used as the reference area MSa. Instead of only one reference area MSa, a plurality of reference areas MSa may be formed on the mesh MS. The shift amount can be calculated with high accuracy using a plurality of reference areas MSa.

FIG. 11 is a recording table in which the shift amount acquired in step S8 is recorded. Similar to the image data D1 and the image data D2, identification information (ID) of each mesh MS and a number (NO) of each holder 24 are associated with information about the shift amount. Such a recording table is stored in the image processing unit C6 or the integrated control unit C0. According to the recording tables in FIGS. 10 and 11, it is possible to manage that the predetermined mesh MS and the holder 24 correspond to which shift amount, image data D1, and image data D2.

Another example of the method for calculating the shift amount includes a method using a screen matching technique. In this case, the image processing unit C6 has the screen matching technique. In the screen matching technique, a shape of the mesh MS in the image data D2 and a shape of the mesh MS in the image data D1 are compared with each other, and a difference between the shapes is calculated as position data, thereby obtaining a shift amount.

In step S9, the charged particle beam device 3 sets the shift amount as an offset value of the stage 36. The shift amount obtained in step S8 is output to the stage control unit C4, and the stage control unit C4 sets coordinates of the stage 36 based on the shift amount. As described above, the integrated control unit C0 controls the control units including the stage control unit C4. Therefore, it can be said that the integrated control unit C0 sets the coordinates of the stage 36 based on the shift amount. Thereafter, a sample is observed by irradiating the sample with an electron beam from the electron gun 31.

As described above, a step of searching for the reference area MSa of the mesh MS is required before observing the sample to be observed using the charged particle beam device 3. Even when the mesh MS can be mounted on the holder 24 with high accuracy, since a position of the mesh MS is shifted when the mesh presser MSH is disposed on the mesh MS, the step of searching for the reference area MSa cannot be omitted.

On the other hand, the coordinates of the stage 36 are set based on the shift amount when a sample is observed in the first embodiment. Therefore, the step of searching for the reference area MSa can be omitted, and a desired sample can be quickly placed in an imaging field of view. That is, observation throughput of the sample can be shortened.

The mesh MS is mounted on the holder 24 with high accuracy in step S2, and when the mounting accuracy of the mesh MS is low, there is no big problem if the technique according to the first embodiment is used. That is, since the shift amount is calculated in step S8 based on the image data D1 acquired in step S2 and the image data D2 acquired in step S4, the stage 36 can be controlled such that a desired sample enters a correct imaging field of view regardless of the mounting accuracy of the mesh MS.

Although step S6 is performed before steps S7 to S9 in the first embodiment, step S6 may be performed after step S7, after step S8, or after step S9. As described above, the identification information (ID) of each mesh MS and the number (NO) of each holder 24 are associated with the information about the shift amount, the image data D1, and the image data D2. Therefore, there is no problem even when a timing at which the holder 24 is fixed to the stage 36 is before or after processing of information such as the shift amount, the image data D1, and the image data D2. That is, it is sufficient that the coordinates of the stage 36 reflect the shift amount before the sample is irradiated with the electron beam.

<<Analysis Function>>

The charged particle beam device 3 has an analysis function for analyzing a sample. After step S9, a sample is observed by the analysis function of the charged particle beam device 3.

An electron beam emitted from the electron gun 31 is focused by the irradiation lens 32 and is emitted onto a sample mounted on the mesh MS. The transmitted electron beam transmitted through the sample is imaged by the objective lens 33, and the captured transmission image is enlarged by the projection lens 34.

When the detector 35 is a fluorescent plate, the captured and enlarged transmission image is projected on the fluorescent plate and recorded as image data in the detector control unit C3 or the integrated control unit C0. The acquired image data is displayed on the display 41, and a user can check a detailed structure of the sample.

Modification

Hereinafter, a modification of the first embodiment will be described.

In step S3 in the first embodiment, the positional relationship between the mesh MS and the holder 24 is imaged using the camera 26, and the captured image is recorded as the image data D1. In the modification, the imaging in step S3 is omitted, and an image captured in advance inside or outside the transfer device 2 is used as the image data D1.

Similar to the first embodiment, the image data D1 in the modification is also image data indicating a state in which the mesh MS is disposed on the holder 24.

For example, as shown in FIG. 1, a plurality of meshes MS are stored in the mesh holding container 22. Samples having the same shape are mounted on these meshes MS. Therefore, the image data D1 obtained by imaging a first mesh MS can be applied as the image data D1 of second and subsequent meshes MS. Therefore, imaging of the second and subsequent meshes MS can be omitted, so that observation throughput can be further shortened.