JP2021107807A - Charged particle beam inspection device and charged particle beam inspection method - Google Patents

Abstract

To provide a charged particle beam inspection device having a small swing width of beam deflection in pattern inspection performed while continuously moving a stage using multi-beam.SOLUTION: An inspection device 100 comprises: a deflector 208 having a first function for collectively deflecting multi-beams on a small region group of N×N' pieces on a substrate in a plurality of small regions having an inspection region of the substrate divided by p/M in an x direction and a predetermined size in a y direction using the multi-beams having a pitch p and N×N' columns to perform tracking deflection and performing tracking reset until the movement of a stage having a distance obtained by N/Mp in a -x direction is completed and a second function for scanning the small region group of N×N' pieces by a first step of repeatedly performing deflection along the y direction while performing the tracking deflection and a second step of repeatedly performing deflection in a -y direction; and a detector 222 for detecting secondary electrons. The value of combinations having the greatest common divisor of 1 is used as the value of N and the value of M.SELECTED DRAWING: Figure 1

Description

The present invention relates to a charged particle beam inspection apparatus and a charged particle beam inspection method. For example, the present invention relates to an inspection device that inspects a pattern by acquiring a secondary electron image of a pattern emitted by irradiating a multi-beam with an electron beam.

In recent years, with the increasing integration and capacity of large-scale integrated circuits (LSIs), the circuit line width required for semiconductor devices has become narrower and narrower. These semiconductor elements use an original image pattern (also referred to as a mask or reticle, hereinafter collectively referred to as a mask) on which a circuit pattern is formed, and the pattern is exposed and transferred onto a wafer by a reduction projection exposure apparatus called a stepper. Manufactured by forming a circuit.

Further, improvement of the yield is indispensable for manufacturing an LSI, which requires a large manufacturing cost. However, immersion exposure and multi-patterning technology have already realized processing dimensions of less than 20 nm, and further practical application of EUV (Extreme Ultraviolet) exposure is about to realize fine processing of less than 10 nm. Further, practical application of microfabrication techniques other than using an exposure machine, such as NIL (NanoPrinting Photolithography) and DSA (Directed Self-Assembly, self-assembling lithography), is also progressing. In recent years, with the miniaturization of LSI pattern dimensions formed on semiconductor wafers, the dimensions that must be detected as pattern defects have become extremely small, and the number of patterns that must be inspected even if they have the same area. Is also enormous. Therefore, it is necessary to improve the accuracy and speed of the inspection device for inspecting the defects of the ultrafine pattern transferred on the semiconductor wafer. In addition, one of the major factors for reducing the yield is a pattern defect of a mask used when exposing and transferring an ultrafine pattern on a semiconductor wafer by photolithography technology. Therefore, it is necessary to improve the accuracy of the inspection device for inspecting the defects of the transfer mask used in LSI manufacturing.

The inspection method includes an optical image obtained by capturing a pattern formed on a sample to be inspected such as a wafer such as a semiconductor wafer or a mask such as a lithography mask using a magnifying optical system at a predetermined magnification, design data, or an inspector. A method of performing an inspection by comparing the same pattern on an inspection sample with an captured optical image is known. For example, as an inspection method, "die to die inspection" in which optical image data obtained by capturing the same pattern in different places on the same mask are compared with each other, or CAD data in which a pattern is designed is drawn on a mask. Sometimes, drawing data (design pattern data) converted into a device input format for input by the drawing device is input to the inspection device, and design image data (reference image) is generated based on this, and this design image data and , There is a "die-to-database (die-database) inspection" that compares an optical image that is measurement data obtained by capturing a pattern. In the inspection method in such an inspection apparatus, the substrate to be inspected is placed on a stage (sample table), and the moving of the stage causes the light flux to scan on the sample to be inspected to perform the inspection. The substrate to be inspected is irradiated with a luminous flux by a light source and an illumination optical system. The light transmitted or reflected through the substrate to be inspected is imaged on the sensor via the optical system. The image captured by the sensor is sent to the comparison circuit as measurement data. In the comparison circuit, after the images are aligned with each other, the measurement data and the reference data are compared according to an appropriate algorithm, and if they do not match, it is determined that there is a pattern defect.

In the pattern inspection apparatus described above, an optical image is acquired by irradiating a substrate to be inspected with a laser beam and capturing a transmitted image or a reflected image thereof. On the other hand, the inspection target substrate is irradiated with a multi-beam composed of a plurality of electron beams in an array arrangement in which a plurality of beam rows arranged at the same pitch on a straight line are arranged and emitted from the inspection target substrate. Development of an inspection device that detects secondary electrons corresponding to each beam and acquires a pattern image is also in progress. In a pattern inspection apparatus using an electron beam including such a multi-beam, secondary electrons are detected by scanning each small area of the substrate to be inspected. At that time, a so-called step-and-repeat operation is performed in which the position of the inspection target substrate is fixed while the beam is being scanned and the position of the inspection target substrate is moved to the next small area after the scanning is completed. By using an array-arranged multi-beam in which multiple beam sequences arranged at the same pitch on a straight line are arranged, a large number of beams can be arranged in a limited area, so that a large number of small areas can be scanned at one time. It will be possible to do it at the same time. Therefore, improvement in throughput is expected. However, in the step-and-repeat operation, a settling time (overhead time) is required for the stage position to stabilize each time the stage is moved. Since the scanning range (small area) at one time is small, the number of steps of the stage becomes enormous in order to scan the entire substrate. Therefore, a wasted time that is not required for scanning is generated by the time obtained by multiplying the number of steps by the settling time. Even when scanning on a substrate using a multi-beam, there is a trial calculation that, for example, a time not required for scanning for 80 hours or more is generated for one substrate.

Therefore, in order to improve the throughput of the inspection device, it is considered to change the stage moving method from the step-and-repeat operation method to the continuous moving method that does not require the settling time for each step. However, when scanning with an array of multi-beams, the continuous movement method can eliminate the need for settling time, but instead, the same small area is sequentially sent within the scanning range of multiple beams arranged in the moving direction. Therefore, useless scanning is repeated for a small area for which a pattern image has already been acquired. Therefore, it does not lead to an improvement in throughput. In order to prevent such useless scanning from being repeated, it is necessary to jump over the already scanned small area and scan the next small area. Therefore, it is necessary to increase the swing width for deflecting the multi-beam. be. However, when the swing width of the beam deflection is increased, the influence of the aberration of the electron optics system becomes large, it becomes difficult to sufficiently narrow each beam to a small size, and so-called blurring occurs.

The following technologies are disclosed as technologies for suppressing this. That is, N rows (N is an integer of 2 or more) at the same pitch p on the substrate surface in the first direction and N'rows (N'is an integer of 1 or more) in the second direction orthogonal to the first direction. Using a multi-beam composed of a plurality of charged particle beams arranged side by side, the inspection area of the substrate has a size obtained in p / M (M is an integer of 2 or more) in the first direction and a predetermined size in the second direction. Of the plurality of divided subregions divided by, the multi-beams are collectively grouped into N × N'divided subregions on the substrate in which N are arranged at a pitch p in the first direction and N'are arranged in the second direction. And bias. While the stage continuously travels the distance obtained by N / M · p in the opposite direction of the first direction, the multi-beam is track-deflected to follow the continuous movement of the stage. Then, while the multi-beam is tracking-deflected so as to follow the continuous movement of the stage, the multi-beam is collectively deflected so as to scan the N × N ′ divided small region group.

According to the above technique, the swing width of the beam deflection in the first direction can be reduced. Therefore, the influence of the aberration of the electron optics system can be reduced.

Here, depending on the type of the sample to be inspected, the pattern image may become white or, conversely, black due to the charging of the sample to be inspected. Therefore, it may be difficult to determine the presence or absence of pattern defects. Therefore, on the sample to be inspected, for example, the multi-beam is scanned in different directions for the above small region (including the divided small region) to acquire a plurality of pattern images, and then the image obtained by averaging the plurality of pattern images is used. It is conceivable to carry out an inspection. Acquisition of such an averaged image can be easily performed when the stage does not move continuously. Next, as described above, consider the case where the stage movement method is the continuous movement method. In this case, when performing a beam scan, there is a problem that an unintended vibration is applied to the stage due to the continuous movement of the stage, and a specific part of the sample to be inspected jumps out of the beam scanable area. In addition, this problem has become particularly large when performing a beam scan at the edge of the above-mentioned small region. Further, the unintended vibration applied to the stage is particularly likely to occur in the moving direction of the stage or the opposite direction of the moving direction of the stage.

Japanese Unexamined Patent Publication No. 2018-017571

Therefore, one aspect of the present invention is to remove the influence of unintended stage vibration and move in different directions in a pattern inspection performed while continuously moving the stage using a multi-beam in which a plurality of beams are arranged in the moving direction of the stage. Provided are an inspection device and an inspection method for averaging images acquired by beam scanning.

The charged particle beam inspection device according to one aspect of the present invention includes a movable stage on which a substrate is placed, a stage control circuit for continuously moving the stage in the direction opposite to the first direction, and a substrate surface in the first direction. Above, it is composed of a plurality of charged particle beams arranged in N rows (N is an integer of 2 or more) and N'rows (N'is an integer of 1 or more) in the second direction orthogonal to the first direction at the same pitch p. The inspection area of the substrate is divided into a plurality of small areas having a size obtained by p / M (M is an integer of 2 or more) in the first direction and a predetermined size in the second direction by using the multi-beam. Among them, the multi-beam is collectively deflected to a small region group of N × N'on a substrate in which N pieces are arranged at a pitch p in the first direction and N'pieces are lined up in the second direction, and the stage is the first stage. While continuously moving the distance obtained by N / M ・ p in the opposite direction of the direction, the multi-beam is tracked and deflected to follow the continuous movement of the stage, and N / M ・ p is moved in the opposite direction of the first direction. By the time the movement of the stage of the distance obtained in is completed, new N × N'pieces arranged at a pitch p in the first direction, which are N pieces away from the N × N'small area group in the first direction. The first function to reset the tracking by re-deflecting the multi-beams to the small area group of the In each of the plurality of subregions, starting from the end on the side opposite to the first direction in each of the plurality of subregions, and in each of the plurality of subregions, the side in the first direction. The batch deflection of the multi-beam along the second direction is repeatedly performed from the end on the side opposite to the first direction toward the end on the side in the first direction, with the end of Step 1 is performed, and then, in each of the plurality of small regions, the end on the side opposite to the first direction is used as the starting point, and the end on the side in the first direction in each of the plurality of small regions. With the portion as the end point, the collective deflection of the multi-beam along the opposite direction of the second direction is repeatedly performed from the end on the opposite side of the first direction to the end on the side in the first direction. By performing the second step, a deflector having two functions of a second function of collectively deflecting the multi-beam so as to scan a small region group of N × N', and a multi-beam on the substrate. It is equipped with a detector that detects secondary electrons emitted from the substrate due to irradiation with, and as the value of N and the value of M, the maximum commitment between the value of N and the value of M is Combination that becomes 1 It is a charged particle beam inspection apparatus characterized by using the value of.

In the above-mentioned charged particle beam inspection apparatus, the measurement pixel size that can be irradiated with the charged particle beam includes PS, the second beam setting time Ofs_v in the second direction or the opposite direction of the second direction, or the second direction or When the beam scan time in the opposite direction of the second direction is Tv, the moving speed V of the stage is preferably V = PS (M-1) / (2Tv).

In the above-mentioned charged particle beam inspection device, when the measurement pixel size that can be irradiated with the charged particle beam is PS and the scan frequency of the charged particle beam is f, the beam in the second direction or the opposite direction of the second direction. The charged particle beam inspection apparatus according to claim 1, wherein the scan time Tv is Tv = (p / PS) × (p / M / PS) × (1 / f) + (p / M / PS) × Ofs_v.

In the above-mentioned charged particle beam inspection apparatus, the average secondary electrons obtained by averaging the first secondary electron image acquired in the first step and the second secondary electron image acquired in the second step. It is preferable to further include an average image acquisition circuit for acquiring images.

It is a block diagram which shows the structure of the pattern inspection apparatus in Embodiment 1. FIG. It is a conceptual diagram which shows the structure of the molded aperture array member in Embodiment 1. FIG. It is a conceptual diagram for demonstrating an example of a scanning operation in Embodiment 1. FIG. It is a figure which shows an example of the irradiation area of the multi-beam and the pixel for measurement in Embodiment 1. FIG. It is a conceptual diagram for demonstrating an example of the detail of a scanning operation in the comparative example of Embodiment 1. FIG. It is a conceptual diagram for demonstrating an example of the details of the scanning operation in Embodiment 1. FIG. It is a figure which shows an example of the relationship between the number of beams and the number of divisions in Embodiment 1. It is a figure which shows the relationship between the sub region and the scanning region in Embodiment 1. FIG. It is a schematic diagram which shows an example of the scanning operation in the 1st aspect of Embodiment 1. It is a schematic diagram which shows an example of the scanning operation in the 1st aspect of Embodiment 1. It is a schematic diagram which shows an example of the scanning operation in the 2nd aspect of Embodiment 1. It is a schematic diagram which shows another example of the scanning operation in the 2nd aspect of Embodiment 1. FIG. It is a schematic diagram which shows another example of the scanning operation in the 2nd aspect of Embodiment 1. FIG. It is a schematic diagram which shows another example of the scanning operation in the 2nd aspect of Embodiment 1. FIG. It is a schematic diagram which shows an example of the scanning operation in the other comparative example of Embodiment 1. It is a schematic diagram which shows an example of the scanning operation in the other comparative example of Embodiment 1. It is a flowchart which shows a part of the main part process of the inspection method in Embodiment 1. FIG. It is a conceptual diagram for demonstrating another example of the detail of the scanning operation in Embodiment 1. FIG. It is a conceptual diagram for demonstrating an example of the relationship between the sub-region and the corresponding beam in the scanning operation in Embodiment 1. FIG. It is a conceptual diagram for demonstrating another example of the scanning operation in Embodiment 1. FIG. It is a figure which shows the internal structure of the comparison circuit in Embodiment 1. FIG. It is a figure which shows an example of the relationship between a grid and a frame area in Embodiment 1. FIG. It is a block diagram which shows the other structure of the pattern inspection apparatus in Embodiment 1. FIG.

Hereinafter, in the embodiment, a case where an electron beam is used as an example of a charged particle beam will be described. However, it is not limited to this. Other charged particle beams such as ion beams may be used.

Embodiment 1.
FIG. 1 is a configuration diagram showing a configuration of a pattern inspection device according to the first embodiment. In FIG. 1, the inspection device 100 for inspecting a pattern formed on a substrate is an example of a charged particle beam inspection device. The inspection device 100 includes an electro-optical image acquisition mechanism 150a and a control system circuit 160 (control unit). The electro-optical image acquisition mechanism 150a includes an electron beam column 102 (electron lens barrel), an examination room 103, a detection circuit 106, a stripe pattern memory 123, and a laser length measuring system 122. In the electron beam column 102, an electron gun (irradiation source) 201, an electromagnetic lens 202, a molded aperture array substrate 203, a reduction lens 205, an electromagnetic lens 206, an electromagnetic lens (objective lens) 207, a deflector 208, and a batch blanking deflection A device 212, a limiting aperture substrate 213, a beam separator 214, a deflector 218, an electromagnetic lens 224, 226, and a multi-detector 222 are arranged.

In the examination room 103, an XY stage (sample table, an example of a stage) 105 that can move at least on an XY plane is arranged. On the XY stage 105, a substrate (sample to be inspected) 101 on which a chip pattern to be inspected is formed is arranged. The substrate 101 is, for example, a silicon wafer or the like. The substrate 101 is arranged on the XY stage 105, for example, with the pattern forming surface facing upward. Further, on the XY stage 105, a mirror 216 that reflects the laser beam for laser length measurement emitted from the laser length measuring system 122 arranged outside the examination room 103 is arranged. The multi-detector 222 is connected to the detection circuit 106 outside the electron beam column 102. The detection circuit 106 is connected to the stripe pattern memory 123.

In the control system circuit 160, the control computer 110, which is a computer, transmits the position circuit 107, the expansion circuit 111, the stage control circuit 114, the lens control circuit 124, the blanking control circuit 126, the deflection control circuit 128, and the image via the bus 120. Storage device 132, comparison circuit 108, storage device 109 such as magnetic disk device, monitor 117, memory 118, printer 119, reference circuit 112, settling time storage device 140, scan frequency storage device 141, average image acquisition circuit 144, moving speed. It is connected to the calculation circuit 146, the scan time calculation circuit 148, and the scan time storage device 149. Further, the XY stage 105 is driven by the drive mechanism 142 under the control of the stage control circuit 114. In the drive mechanism 142, for example, a drive system such as a three-axis (XY−θ) motor that drives in the X direction, the Y direction, and the θ direction is configured, and the XY stage 105 can be moved. As these X motors, Y motors, and θ motors (not shown), for example, step motors can be used. The XY stage 105 can be moved in the horizontal direction and the rotational direction by a motor of each axis of XYθ. Then, the moving position of the XY stage 105 is measured by the laser length measuring system 122 and supplied to the position circuit 107. The laser length measuring system 122 measures the position of the XY stage 105 by receiving the reflected light from the mirror 216 based on the principle of the laser interferometry method.

A high-pressure power supply circuit (not shown) is connected to the electron gun 201, and an acceleration voltage from the high-pressure power supply circuit is applied between the filament and the extraction electrode (not shown) in the electron gun 201, and a predetermined voltage of the extraction electrode is applied and predetermined. By heating the cathode (filament) at the temperature of, the group of electrons emitted from the cathode is accelerated and emitted as an electron beam. For the reduction lens 205 and the objective lens 207, for example, an electromagnetic lens is used, and both are controlled by the lens control circuit 124. The beam separator 214 is also controlled by the lens control circuit 124. The collective blanking deflector 212 is composed of a group of electrodes having at least two poles and is controlled by the blanking control circuit 126. The deflector 208 is composed of a group of electrodes having at least four poles, and is controlled by the deflection control circuit 128.

When the substrate 101 is a semiconductor wafer on which a plurality of chip (die) patterns are formed, the pattern data of the chip (die) patterns is input from the outside of the inspection device 100 and stored in the storage device 109. When the substrate 101 is an exposure photomask, design pattern data that is a basis for forming a mask pattern on the exposure photomask is input from the outside of the inspection device 100 and stored in the storage device 109.

Here, FIG. 1 describes a configuration necessary for explaining the present embodiment. The inspection apparatus 100 may usually have other necessary configurations.

FIG. 2 is a conceptual diagram showing the configuration of the molded aperture array member according to the first embodiment. In FIG. 2, the molded aperture array substrate 203 has a two-dimensional (matrix) horizontal (x direction) N row × vertical (y direction) N'stage (N is an integer of 2 or more, N'is 1 or more. Holes (openings) 22 (integers) are formed in the x and y directions (x: first direction, y: second direction) with a predetermined arrangement pitch L. The reduction magnification of the multi-beam is a times (when the multi-beam diameter is reduced to 1 / a and the substrate 101 is irradiated), and the pitch between the beams of the multi-beam in the x and y directions on the substrate 101 is p. In this case, the arrangement pitch L has a relationship of L = (a × p). In the example of FIG. 2, a case where 5 × 5 holes 22 for forming a multi-beam of N = 5 and N'= 5 are formed is shown. Next, the operation of the electro-optical image acquisition mechanism 150a in the inspection device 100 will be described.

The electron beam 200 emitted from the electron gun 201 (emission source) is illuminated by the electromagnetic lens 202 substantially vertically on the entire molded aperture array substrate 203. As shown in FIG. 2, a plurality of holes 22 (openings) are formed in the molded aperture array substrate 203, and the electron beam 200 illuminates an area including all the plurality of holes 22. A multi-primary electron beam (multi-beam) 20 is formed by passing each part of the electron beam 200 irradiated to the positions of the plurality of holes 22 through the plurality of holes 22 of the molded aperture array substrate 203. NS.

The formed multi-beam 20 is deflected by an electromagnetic lens (reduction lens) 205 and an electromagnetic lens 206, respectively, and is arranged at a crossover position of each beam of the multi-beam 20 while repeating an intermediate image and a crossover. It passes through the separator 214 and proceeds to the objective lens 207. Then, the objective lens 207 focuses the multi-beam 20 on the substrate 101. The multi-beam 20 focused (focused) on the surface of the substrate 101 (sample) by the objective lens 207 is collectively deflected by the deflector 208 and irradiates each irradiation position on the substrate 101 of each beam. Will be done. When the entire multi-beam 20 is collectively deflected by the batch blanking deflector 212, the position is displaced from the central hole of the limiting aperture substrate 213 and is shielded by the limiting aperture substrate 213. On the other hand, the multi-beam 20 not deflected by the batch blanking deflector 212 passes through the central hole of the limiting aperture substrate 213 as shown in FIG. By turning ON / OFF of the batch blanking deflector 212, blanking control is performed, and ON / OFF of the beam is collectively controlled. In this way, the limiting aperture substrate 213 shields the multi-beam 20 deflected so that the beam is turned off by the batch blanking deflector 212. Then, the multi-beam 20 for inspection (for image acquisition) is formed by the beam group that has passed through the limiting aperture substrate 213 formed from the time when the beam is turned on to the time when the beam is turned off.

When the multi-beam 20 is irradiated to a desired position on the substrate 101, secondary electrons including backscattered electrons corresponding to each beam of the multi-beam 20 from the substrate 101 due to the irradiation of the multi-beam 20 A bundle (multi-secondary electron beam 300) is emitted.

The multi-secondary electron beam 300 emitted from the substrate 101 passes through the objective lens 207 and proceeds to the beam separator 214.

Here, the beam separator 214 generates an electric field and a magnetic field in a direction orthogonal to each other on a plane orthogonal to the direction in which the central beam of the multi-beam 20 travels (the central axis of the electron orbit). The electric field exerts a force in the same direction regardless of the traveling direction of the electron. On the other hand, the magnetic field exerts a force according to Fleming's left-hand rule. Therefore, the direction of the force acting on the electron can be changed depending on the invasion direction of the electron. The force due to the electric field and the force due to the magnetic field cancel each other out to the multi-beam 20 that enters the beam separator 214 from above, and the multi-beam 20 travels straight downward. On the other hand, in the multi-secondary electron beam 300 that invades the beam separator 214 from below, both the force due to the electric field and the force due to the magnetic field act in the same direction, and the multi-secondary electron beam 300 is obliquely upward. It is bent and separated from the multi-beam 20.

The multi-secondary electron beam 300, which is bent obliquely upward and separated from the multi-beam 20, is further bent by the deflector 218 and projected onto the multi-detector 222 while being refracted by the electromagnetic lenses 224 and 226. The multi-detector 222 detects the projected multi-secondary electron beam 300. Backscattered electrons and secondary electrons may be projected onto the multi-detector 222, or the backscattered electrons may be diverged on the way and the remaining secondary electrons may be projected. The multi-detector 222 has, for example, a two-dimensional sensor (not shown). Then, each secondary electron of the multi-secondary electron beam 300 collides with the corresponding region of the two-dimensional sensor to generate electrons, and secondary electron image data is generated for each pixel. The intensity signal detected by the multi-detector 222 is output to the detection circuit 106.

FIG. 3 is a conceptual diagram for explaining an example of the scanning operation according to the first embodiment. In FIG. 3, the inspection area 30 of the substrate 101 is virtually divided into a plurality of strip-shaped stripe areas 32 having a predetermined width in the y direction, for example. As the substrate 101, it is preferable to apply it to, for example, an exposure mask substrate. For example, it is virtually divided into a plurality of strip-shaped stripe regions 32 having the same width as the natural number of the width of the irradiation region 34 that can be irradiated by one irradiation of the entire multi-beam 20. In the example of FIG. 3, it is virtually divided into a plurality of strip-shaped stripe regions 32 having the same width as the irradiation region 34. First, the XY stage 105 is moved so that the multi-beam 20 is moved once from the left end of the first stripe region 32 to a position outside the first stripe region 32, for example, one size of the irradiation region 34. The tracking area 33 is adjusted so that the irradiation area 34 that can be irradiated by irradiation is located, and the scanning operation is started. In the first embodiment, the XY stage 105 is continuously moved in the −x direction (an example in the direction opposite to the first direction) at a constant velocity, for example, and the irradiation region 34 is moved so as to follow the continuous movement as desired. The tracking reset is performed by scanning the sub-region group arranged at the pitch p in the tracking region 33 and moving the irradiation region 34 to the next tracking region 33 in the x direction (an example of the first direction) after the completion. By repeating this operation, the stripe region 32 is sequentially scanned in the x direction. When scanning the first stripe region 32, the scanning operation is relatively advanced in the x direction by moving the XY stage 105 in the −x direction, for example. When the multi-beam irradiation of the first stripe area 32 is completed, the stage position is moved in the −y direction to a position further to the right of the irradiation area 34, for example, by one size, from the right end of the second stripe area 32. The irradiation region 34 is adjusted so as to be relatively located in the y direction, and this time, by moving the XY stage 105 in the x direction, for example, multi-beam irradiation is performed in the same manner in the −x direction. In the third stripe region 32, multi-beam irradiation is performed in the x direction, in the fourth stripe region 32, multi-beam irradiation is performed in the −x direction, and so on. This can shorten the inspection time. However, the scanning is not limited to the case where the scanning is performed while alternately changing the directions, and when drawing each stripe region 32, the scanning may be advanced in the same direction. The multi-beam 20 formed by passing through each hole 22 of the molded aperture array substrate 203 allows the multi-beam 20 to be a bundle of secondary electrons corresponding to a plurality of beams (primary electron beams) having the same number as each hole 22 at the maximum. The secondary electron beam 300 is detected at the same time.

FIG. 4 is a diagram showing an example of the irradiation region of the multi-beam and the measurement pixel in the first embodiment. In FIG. 4, each stripe region 32 is divided into a plurality of mesh regions having a multi-beam beam size, for example. Each such mesh region becomes a measurement pixel 36 (unit irradiation region). Then, in the irradiation region 34, a plurality of measurement pixels 28 (irradiation positions of the beams at the time of one shot painted in black) that can be irradiated by one irradiation of N × N'multi-beams 20 are shown. There is. In other words, the pitch p between the adjacent measurement pixels 28 in the x and y directions is the pitch between each beam of the multi-beam 20 on the substrate 101. In the example of FIG. 4, one of the four adjacent measurement pixels 28 is set as one of the four corners of the rectangle, and is surrounded by p × p in the x and y directions starting from the measurement pixel 28. The area (hereinafter referred to as p × p area 27) is divided by the number of divisions M (M is an integer of 2 or more) in the x direction, and is a rectangular area with a size of p / M in the x direction and p in the −y direction. One grid 29 (sub-area; small area) is formed. In the example of FIG. 4, each grid 29 (individual beam scan area) shows a case composed of 3 × 9 pixels.

FIG. 5 is a conceptual diagram for explaining an example of details of the scanning operation in the comparative example of the first embodiment. In the example of FIG. 5, as a comparative example of the first embodiment, among the N × N'multi-beams 20, N multi-beams for one stage in the y direction are shown. The “stage” is the number of grids 29 or p × p regions arranged in the y direction. For example, "one step" means that "the grid 29 or the p × p region is lined up by one in the y direction". Here, N = 5 multi-beams arranged in the x direction at the same pitch p are shown. In the comparative example of the first embodiment, each beam of N = 5 multi-beams arranged in the x direction at the same pitch p is surrounded by p × p in the x and y directions starting from the measurement pixel 28 of the beam. The case where the p × p region 27 surrounded by the next p × p is scanned after scanning the entire p × p region 27 is shown. In the comparative example of the first embodiment, so that each beam XY stage 105 during (t = duration of t _{0 '~t} 1') for scanning a region surrounded by the p × p moves by N · p Control the stage speed. At that time, tracking deflection is performed by the deflector 208 so that each beam can scan the region surrounded by the p × p by the deflection operation of the deflector 208. Then, at the time when the scanning of the p × p region 27 surrounded by N consecutive p × ps arranged in the x direction is completed (t = t ₁ ′), N = 5 so that the scanning regions do not overlap. Tracking reset is performed by deflecting the multi-beams in the x direction all at once. By repeating this operation, the regions on the continuously moving stage can be scanned by the multi-beam so that the scanning regions do not overlap. In the example of FIG. 5, it is necessary to deflect the multi-beam in the x direction (or −x direction) by (N-1) · p (= 4p). Therefore, in the comparative example of the first embodiment, the swing width of the beam deflection in the x direction (or −x direction) is required by (N-1) · p. On the other hand, the swing width of the beam deflection in the y direction (or −y direction) needs to be p. When the number of beams N increases, the swing width of the beam deflection becomes very large. Therefore, as described above, the influence of the aberration of the electron optics system becomes large.

FIG. 6 is a conceptual diagram for explaining an example of details of the scanning operation according to the first embodiment. In the example of FIG. 6, as the first embodiment, of the N × N ′ multi-beams 20, N multi-beams for one stage in the y direction are shown. Here, as in FIG. 5, N = 5 multi-beams arranged in the x direction at the same pitch p are shown. In the first embodiment, one of the four adjacent measurement pixels 28 is set as one of the four corners of the rectangle and is surrounded by p × p in the x and y directions starting from the measurement pixel 28. The p × p region 27 is divided in the x direction by the number of divisions M. Therefore, one grid 29 (sub-region; small region) is composed of a rectangular region having a size of p / M in the x direction and p in the −y direction (predetermined size). In the example of FIG. 6, the case where the number of divisions M = 3 is shown. In the first embodiment, each beam of N = 5 multi-beams arranged in the x direction at the same pitch p has a size of p / M in the x direction and p in the y direction starting from the measurement pixel 28 of the beam. The case where the next grid 29 separated by N pieces in the x direction is scanned after scanning the grid 29 (sub-region) of a predetermined size) is shown.

In FIG. 6, in the first embodiment, it is the same as the comparative example of FIG. 5 while each beam scans the grid 29 surrounded by (p / M) × p ( _{period from t = t 0 to t 1).} In the case of stage speed, the XY stage 105 moves by N / M · p. At that time, N lines are arranged at a pitch p in the x direction so that each beam can scan the grid 29 surrounded by the (p / M) × p by the deflection operation by the second function of the deflector 208 (p / M). With the grid 29 having a size of M) × p as the tracking region 33, the deflector 208 performs tracking deflection by the first function. Then, by the second function of the deflector 208, when the scanning of the grid 29 having a size of (p / M) × p in which N lines are arranged at a pitch p in the x direction is completed (t = t ₁ ), the scanning area is set. The deflector 208 uses the first function to perform tracking reset by collectively deflecting N = 5 multi-beams to positions separated by N grids by 29 minutes in the x direction so as not to overlap. .. In the example of FIG. 6, the deflector 208 uses the first function to collectively deflect five multi-beams at positions separated by the five grids 29. At that time, it goes without saying that the deflection position by the second function of the deflector 208 is reset from the last pixel 36 in the grid 29 to the first pixel 28. t ₌ t 1 ~t ₂ period, _{t =} duration of t 2 ~t _3, and ..., by repeating such operation, does not overlap the scanning area on the same stripe region 32 even if for continuously moving the stage Can be scanned with multiple beams. In the example of FIG. 6, it is necessary to deflect the multi-beam in the x direction by (N-1) / M · p (= 4 p / M). Therefore, in the first embodiment, the swing width of the beam deflection in the x direction can be suppressed to (N-1) / M · p. However, if the relationship between the number of beams N in the x direction and the number of divisions M is not controlled, scanning omission (toothlessness) or grid 29 (sub-region) of overlapping scanning will occur. In the first embodiment, in order to apply such a scanning method, a value of a combination in which the greatest common divisor between the number of beams N in the x direction and the number of divisions M is 1 is used. Under such conditions, scanning omission (toothlessness) or duplicate scanning can be avoided.

FIG. 7 is a diagram showing an example of the relationship between the number of beams and the number of divisions in the first embodiment. FIG. 7 shows a scanning operation when the number of divisions M is changed by using N = 7 beams in the x direction. Further, in FIG. 7, each time the tracking reset is performed, the beam is shown as scanning another stage for the sake of clarity in the illustration. In FIG. 7, for convenience, the size of the p × p region 27 surrounded by p × p in the y direction is narrowed. FIG. 7A shows a case where the number of divisions M = 1, that is, the p × p region 27 surrounded by p × p is not divided. In FIG. 7A, when the tracking reset is performed, the swing width of the beam deflection becomes as large as 6p. FIG. 7B shows a case where the number of divisions M = 2, that is, the p × p region 27 surrounded by p × p is divided into two. In FIG. 7B, when the tracking reset is performed, the swing width of the beam deflection can be reduced to 3p. FIG. 7C shows a case where the number of divisions M = 3, that is, the p × p region 27 surrounded by p × p is divided into three. In FIG. 7C, when the tracking reset is performed, the swing width of the beam deflection can be reduced to 2p. FIG. 7D shows a case where the number of divisions M = 4, that is, the p × p region 27 surrounded by p × p is divided into four. In FIG. 7D, when the tracking reset is performed, the swing width of the beam deflection can be reduced to (3/2) p. FIG. 7E shows a case where the number of divisions M = 5, that is, the p × p region 27 surrounded by p × p is divided into five. In FIG. 7 (e), when the tracking reset is performed, the swing width of the beam deflection can be reduced to (6/5) p. FIG. 7 (f) shows a case where the number of divisions M = 6, that is, the p × p region 27 surrounded by p × p is divided into six. In FIG. 7 (f), when the tracking reset is performed, the swing width of the beam deflection can be reduced to p. By increasing the number of divisions M in this way, the swing width of the beam deflection can be made smaller.

Here, when one beam scans a sub-region (grid 29) in which the p × p region 27 surrounded by p × p is divided into M, the multi-beam 20 having the number of beams N is scanned in the x direction. Then, N sub-regions (grid 29) arranged for every M pieces are scanned at the same time. Here, M × N continuous sub-regions (grid 29) groups are set as one predetermined range. When the first beam of the multi-beam 20 moves in the x direction by one predetermined range, the sub-region that failed to be scanned remains as it is without being scanned. Here, assuming that the number (movement amount) of jumping over the sub-region when performing tracking reset is D, M × N / while the first beam of the multi-beam 20 moves in the x direction by one predetermined range. The tracking cycle operation will be performed D times. Therefore, in order to scan all the sub-regions that were scanned only once for each M without duplication and omission, the number of divisions M and the number of tracking cycle operations are the same, that is, M =. Must be M × N / D. Therefore, D = N. Therefore, in the first embodiment, the number D that jumps over the sub-region when the tracking reset is performed has the same value as the number N of beams in the x direction. The swing width of the beam at that time is (N-1) p / M.

When N sub-areas (grid 29) arranged for every M pieces are scanned at the same time and the number of jumps over the sub-areas when performing tracking reset is set to N, the scanning ranges do not overlap in one predetermined range. In order to do so, the following relationship is necessary.
0, M, 2M, 3M, ..., (N-1) M, NM
0, N, 2N, 3N, ..., (M-1) N, MN

It is necessary to prevent these two sequences from having the same value in the middle. Therefore, a combination value (a relatively prime relationship between the number of beams N and the number of divisions M) is required so that the greatest common divisor between the number of beams N and the number of divisions M in the x direction is 1. In the examples of FIGS. 7 (a) to 7 (f), when the number of divisions M = 7, the same value is obtained in the middle. Specifically, when performing a tracking reset, the sub-region after movement is already scanned by the adjacent beam, so it overlaps and is NG.

Further, as the value of the number of beams N, as shown in the examples of FIGS. 7 (a) to 7 (f), it is more preferable to use a prime number. By setting the number of beams N to a prime number (for example, 2,3,5,7,11,13,17,23, ...), the degree of freedom of the number of divisions M can be dramatically increased.

FIG. 8 is a diagram showing the relationship between the sub-region 29 and the scanning region in the first embodiment. FIG. 8 shows a diagram in which the region surrounded by the beam pitch p × p is divided into three in the x direction. As shown in FIG. 8, when each beam of the multi-beam 20 scans the corresponding sub-region (grid 29), each beam overlaps (overlaps) a part of the adjacent sub-region (grid 29). It is preferable to set the scanning area 31 of. The adjacent sub-regions (grid 29) are scanned by different beams of the multi-beam 20. Therefore, the inter-beam pitch p deviates from the equal pitch due to the influence of the aberration of the optical system. Therefore, it is preferable to provide a margin width α capable of absorbing such a deviation. Therefore, it is preferable that the scanning region 31 for actual scanning is a region in which a margin width α is added to at least one end side of both ends in the x direction with respect to the sub region (grid 29). By adding such a margin, the scanning end position moves in the x direction by the margin width α. If the margin widths α are provided on both sides, the scanning start position also moves in the −x direction by the margin width α. FIG. 8 shows a case where the margin width α is added in the x direction, but it is more preferable to add the margin width α in the y direction as well.

As described with reference to FIG. 7, the swing width of the beam can be reduced by increasing the number of divisions M. Therefore, from the viewpoint of reducing the swing width of the beam, it is desirable that the number of divisions is large. On the other hand, if the number of divisions M is increased, the number of sub-regions (grid 29) increases, so the number of overlapping portions increases, and useless image data increases. This increases the amount of data. Therefore, from the viewpoint of reducing the amount of data, it is desirable that the number of divisions is small. Therefore, it is more preferable to select the minimum value of the number of divisions M that can obtain the swing width of the beam in which the influence of the aberration of the electron optics system can be ignored.

FIG. 9 is a schematic diagram showing an example of a scanning operation according to the first aspect of the first embodiment. The actual scan is performed by scanning the beam for each measurement pixel 28 as shown in FIG. 4, but here, for the sake of understanding, the scanning operation is schematically shown by an arrow. In FIG. 9, the x direction, which is the right direction of the paper surface, is the first direction, the −x direction, which is the left direction of the paper surface, is the opposite direction of the first direction (stage moving direction), and the y direction, which is the upward direction of the paper surface, is the first direction. The direction 2 and the -y direction, which is the downward direction on the paper, are the opposite directions of the second direction.

FIG. 9A shows the first step of the scanning operation. In the first step, the beam is deflected along the y direction. Here, the deflection of the beam in the first step starts from the end 29a on the −x direction side of the grid 29 (for example, the bottom of the end 29a), and the end on the x direction side of the grid 29. This is performed with the portion 29b (for example, the uppermost portion of the end portion 29b) as an end point. Further, the deflection of the beam in the first step is repeatedly performed from the end portion 29a on the −x direction side of the grid 29 toward the end portion 29b on the x direction side of the grid 29. Therefore, as an example of scanning the beam in FIG. 9 (a), it is first scanned from (1) to (2), then from (3) to (4), and then from (5). It is scanned to (6). For example, the scan from (1) to (2) is a scan of the pixel 28 provided at the end portion 29a on the −x direction side in the grid 29. For example, the scan from (5) to (6) is a scan of pixels 28 provided at the end portion 29b on the x-direction side in the grid 29. The scan from (3) to (4) is a scan of the pixels 28 provided between the end portion 29a on the −x direction side and the end portion 29b on the x direction side in the grid 29. Since the XY stage 105 continues to move in the −x direction during the series of scans, the grid 29 also continues to move in the −x direction.

FIG. 9B shows a second step of the scanning operation. In the second step, the beam is deflected along the −y direction. Here, the deflection of the beam in the second step starts from the end 29a on the −x direction side of the grid 29 (for example, the uppermost portion of the end 29a), and the end on the x direction side of the grid 29. This is performed with the portion 29b (for example, the lowermost portion of the end portion 29b) as an end point. Further, the deflection of the beam in the second step is repeatedly performed from the end portion 29a on the −x direction side of the grid 29 toward the end portion 29b on the x direction side of the grid 29. Therefore, as an example of scanning the beam in FIG. 9B, it is first scanned from (19) to (20), then from (21) to (22), and then from (23). Scanned to (24). For example, the scan from (19) to (20) is a scan of pixels 28 provided at the end 29a on the −x direction side in the grid 29. For example, the scan from (23) to (24) is a scan of pixels 28 provided at the end 29b on the x-direction side in the grid 29. The scan from (21) to (22) is a scan of the pixels 28 provided between the end portion 29a on the −x direction side and the end portion 29b on the x direction side in the grid 29. Since the XY stage 105 continues to move in the −x direction during the series of scans, the grid 29 also continues to move in the −x direction.

The scanning operation is performed by sequentially performing the first step and the second step described above.

FIG. 10 is a schematic view showing another example of the scanning operation in the first aspect of the first embodiment. It is different from FIG. 9 in that the y direction, which is the upward direction on the paper surface, is the opposite direction of the second direction, and the −y direction, which is the downward direction on the paper surface, is the second direction. Further, in FIG. 10A, the point that the beam is deflected along the −y direction is different from that in FIG. 9A. Further, in FIG. 10B, the point that the beam is deflected along the + y direction is different from that in FIG. 9B.

When the beam scan time in the y direction or −y direction including the second beam setting time Ofs_v in the y direction or −y direction is Tv, the moving speed V of the XY stage 105 is V = PS (M-1) /. (2Tv). This is derived as follows. First, the area of the grid 29 in charge of one beam is (p / M) × p. The beam scan is started after the grid 29 having (p / M) × p completely enters the p × p region 27 surrounded by p × p in the x direction and the y direction. In order to minimize the influence of the beam aberration on the image, it is preferable to complete the scan within the p × p region 27 surrounded by p × p. Then, the distance that the XY stage 105 moves from the start to the completion of the beam scan is pp / M = p (1-1 / M). Next, the size of the measurement pixel 28 in the x-direction and the y-direction is defined as PS, and the time required for beam scanning in the y-direction is defined as Tv. In this case, the scan time in the y direction is calculated by (p / (M × PS)) × Tv. Then, as shown in FIGS. 9 and 10, since the polarity is changed in the y direction (y direction and −y direction) and the scan is performed twice, ((p / (M × PS)) × Tv × 2 ) × V = p (1-1 / M). Therefore, V = PS (M-1) / (2Tv)). The moving speed V of the XY stage 105 described above is calculated by, for example, the moving speed calculation circuit 146.

Next, the beam scan time Tv in the y direction or −y direction is Tv = (p / PS) × (p / M / PS) × (1 / f) + (p / M / PS) × Ofs_v. Here, f is a scan frequency used for beam scanning. (P / PS) × (p / M / PS) is the number of measurement pixels 28 in the grid 29. Therefore, the term (p / PS) × (p / M / PS) × (1 / f) corresponds to the time required to irradiate each measurement pixel 28 with a beam in the grid 29. be. Next, (p / M / PS) × Ofs_v is the term for the time it takes to return the beam to scan in the same direction. Therefore, as the sum of these, Tv = (p / PS) × (p / M / PS) × (1 / f) + (p / M / PS) × Ofs_v is derived. The scan frequency is stored in, for example, the scan frequency storage device 141.

FIG. 11 is a schematic diagram showing an example of a scanning operation according to the second aspect of the first embodiment. In FIG. 11, the x direction, which is the right direction of the paper surface, is the first direction, the −x direction, which is the left direction of the paper surface, is the opposite direction of the first direction (stage moving direction), and the y direction, which is the upward direction of the paper surface, is the first direction. The direction 2 and the -y direction, which is the downward direction on the paper, are the opposite directions of the second direction.

FIG. 11A shows the first step of the scanning operation. In the first step, the beam is deflected along the y direction. Here, the deflection of the beam in the first step starts from the end 29a on the −x direction side of the grid 29 (for example, the bottom of the end 29a), and the end on the x direction side of the grid 29. This is performed with the portion 29b (for example, the uppermost portion of the end portion 29b) as an end point. Further, the deflection of the beam in the first step is repeatedly performed from the end portion 29a on the −x direction side of the grid 29 toward the end portion 29b on the x direction side of the grid 29. Therefore, as an example of scanning the beam in FIG. 11 (a), it is first scanned from (1) to (2), then from (3) to (4), and then from (5). It is scanned to (6). For example, the scan from (1) to (2) is a scan of the pixel 28 provided at the end portion 29a on the −x direction side in the grid 29. For example, the scan from (5) to (6) is a scan of pixels 28 provided at the end portion 29b on the x-direction side in the grid 29. The scan from (3) to (4) is a scan of the pixels 28 provided between the end portion 29a on the −x direction side and the end portion 29b on the x direction side in the grid 29. Since the XY stage 105 continues to move in the −x direction during the series of scans, the grid 29 also continues to move in the −x direction.

FIG. 11B shows a second step of the scanning operation. In the second step, the beam is deflected along the −x direction. Here, the deflection of the beam in the second step starts from the end 29c on the −y direction side of the grid 29 (for example, the right end of the end 29c), and the end on the y direction side of the grid 29. This is performed with 29d (for example, the left end of the end 29d) as the end point. Further, the deflection of the beam in the second step is repeatedly performed from the end portion 29c on the −y direction side of the grid 29 toward the end portion 29d on the y direction side of the grid 29. Therefore, as an example of scanning the beam in FIG. 11 (b), it is first scanned from (7) to (8), then from (9) to (10), and then from (11). It is scanned to (12). Since the XY stage 105 continues to move in the −x direction during the series of scans, the grid 29 also continues to move in the −x direction.

FIG. 11C shows a third step of the scanning operation. In the third step, the beam is deflected along the x direction. Here, the deflection of the beam in the second step starts from the end 29c on the −y direction side of the grid 29 (for example, the left end of the end 29c), and the end on the y direction side of the grid 29. This is performed with 29d (for example, the right end of the end 29d) as the end point. Further, the deflection of the beam in the third step is repeatedly performed from the end portion 29d on the y-direction side of the grid 29 to the end portion 29c on the −y direction side of the grid 29. Therefore, as an example of scanning the beam in FIG. 11 (c), it is first scanned from (13) to (14), then from (15) to (16), and then from (17). It is scanned to (18). Since the XY stage 105 continues to move in the −x direction during the series of scans, the grid 29 also continues to move in the −x direction.

FIG. 11D shows a fourth step of the scanning operation. In the fourth step, the beam is deflected along the −y direction. Here, the deflection of the beam in the fourth step starts from the end 29a on the −x direction side of the grid 29 (for example, the uppermost portion of the end 29a), and the end on the x direction side of the grid 29. This is performed with the portion 29b (for example, the lowermost portion of the end portion 29b) as an end point. Further, the deflection of the beam in the fourth step is repeatedly performed from the end portion 29a on the −x direction side of the grid 29 toward the end portion 29b on the x direction side of the grid 29. Therefore, as an example of scanning the beam in FIG. 11 (d), it is first scanned from (19) to (20), then from (21) to (22), and then from (23). Scanned to (24). For example, the scan from (19) to (20) is a scan of pixels 28 provided at the end 29a on the −x direction side in the grid 29. For example, the scan from (23) to (24) is a scan of pixels 28 provided at the end 29b on the x-direction side in the grid 29. The scan from (21) to (22) is a scan of the pixels 28 provided between the end portion 29a on the −x direction side and the end portion 29b on the x direction side in the grid 29. Since the XY stage 105 continues to move in the −x direction during the series of scans, the grid 29 also continues to move in the −x direction.

The scanning operation is performed by sequentially performing the first step, the second step, the third step, and the fourth step described above.

FIG. 12 is a schematic view showing another example of the scanning operation in the second aspect of the first embodiment. In FIGS. 11 (b) and 11 (c), the scan is repeated from the end 29c on the −y direction side of the grid 29 to the end 29d on the y direction side of the grid 29, whereas FIG. 12 (b) and FIG. 12 (c) are different in that the process is repeated from the end 29d on the y-direction side of the grid 29 to the end 29c on the −y direction side of the grid 29.

FIG. 13 is a schematic view showing another example of the scanning operation in the second aspect of the first embodiment. It is different from FIG. 11 in that the y direction, which is the upward direction on the paper surface, is the opposite direction of the second direction, and the −y direction, which is the downward direction on the paper surface, is the second direction. Further, in FIG. 13A, the point that the beam is deflected along the −y direction is different from that in FIG. 11A. Further, in FIG. 13 (d), the point that the beam is deflected along the + y direction is different from that in FIG. 11 (d).

FIG. 14 is a schematic view showing another example of the scanning operation in the second aspect of the first embodiment. In FIGS. 13 (b) and 13 (c), scanning is repeated from the end 29c on the −y direction side of the grid 29 to the end 29d on the y direction side of the grid 29, whereas FIG. 14 (b) and FIG. 14 (c) are different in that scanning is repeatedly performed from the end 29d on the y-direction side of the grid 29 to the end 29c on the −y direction side of the grid 29.

The beam scan time in the x or −x direction, including the first beam settling time Ofs_h in the x or −x direction, includes the second beam settling time Ofs_v in the Th, y or −y direction, y or −y. When the beam scan time in the direction is Tv, the moving speed V of the XY stage 105 is V = PS (M-1) / (2 × (M × Th + Tv)). This is derived as follows. First, the area of the grid 29 in charge of one beam is (p / M) × p. The beam scan is started after the grid 29 having (p / M) × p completely enters the p × p region 27 surrounded by p × p in the x direction and the y direction. In order to minimize the influence of the beam aberration on the image, it is preferable to complete the scan within the p × p region 27 surrounded by p × p. Then, the distance that the XY stage 105 moves from the start to the completion of the beam scan is pp / M = p (1-1 / M). Next, the size of the measurement pixel 28 in the x-direction and the y-direction is PS, the time required for the beam scan in the x-direction is Th, and the time required for the beam scan in the y-direction is Tv. In this case, the scan time in the x direction is calculated by (p / PS) × Th. The scan time in the y direction is calculated by (p / (M × PS)) × Tv. Then, as shown in FIGS. 11 to 14, the polarity is changed in the x direction (x direction and −x direction), and the scan is performed twice, and the polarity is changed in the y direction (y direction and −y direction). Since the scan is performed twice, the formula ((p / (M × PS)) × Tv × 2 + (p / PS) × Th × 2) × V = p (1-1 / M) is established. Therefore, V = PS (M-1) / (2 × (M × Th + Tv)). The moving speed V of the XY stage 105 described above is calculated by, for example, the moving speed calculation circuit 146.

Next, the beam scan time Th in the x-direction or −x direction is Th = (p / PS) × (p / M / PS) × (1 / f) + (p / PS) × Ofs_h. Here, f is a scan frequency used for beam scanning. (P / PS) × (p / M / PS) is the number of measurement pixels 28 in the grid 29. Therefore, the term (p / PS) × (p / M / PS) × (1 / f) corresponds to the time required to irradiate each measurement pixel 28 with a beam in the grid 29. be. Next, (p / PS) × Ofs_h is a term for the time it takes to return the beam to scan in the same direction. Therefore, Th = (p / PS) × (p / M / PS) × (1 / f) + (p / PS) × Ofs_h is derived as the sum of these. The scan frequency is stored in, for example, the scan frequency storage device 141.

Similarly, the beam scan time Tv in the y-direction or −y-direction is Tv = (p / PS) × (p / M / PS) × (1 / f) + (p / M / PS) × Ofs_v.

The first beam settling time Ofs_h and the second beam settling time Ofs_v are stored in the settling time storage device 140. The first beam settling time Ofs_h and the second beam settling time Offs_v may be input by the operator using the control computer 110 and stored in the settling time storage device 140, for example. Further, the beam scan time Th and the beam scan time Tv are calculated by, for example, the scan time calculation circuit 148 based on the above equations. The calculated beam scan time Th and beam scan time Tv may be stored in, for example, the scan time storage device 149.

FIG. 15 is a schematic diagram showing an example of a scanning operation in another comparative example of the first embodiment.

FIG. 15A is for scanning the beam along the −x direction in the first step as shown in FIG. 11A, for example. In this case, in particular, in the scan at the end portion 29b on the x-direction side of the grid 29 as in (1), (3) and (5) of FIG. 15A, the grid 29 is the tracking area 33. If the XY stage 105 jumps out in the −x direction corresponding to the moving direction, the beam is not irradiated to the end portion 29b of the grid 29 on the x direction side, so that the inspection cannot be performed. On the other hand, FIG. 15B scans the beam along the x direction. In this case, in particular, in the scan at the end portion 29a on the −x direction side of the grid 29 as in (2), (4) and (6) of FIG. 15 (b), the grid 29 is the tracking area. If the XY stage 105 protrudes from the 33 in the −x direction corresponding to the moving direction, the beam is not irradiated to the end portion 29a of the grid 29 on the −x direction side, so that the inspection cannot be performed. In order to avoid this problem, it is preferable to perform a beam scan along the y direction or the −y direction in the first step.

FIG. 15 (c) scans the beam along the −x direction in a fourth step, for example, as shown in FIG. 11 (d). In this case, especially in scanning at the end 29a on the −x direction side of the grid 29, such as (20), (22) and (24), the grid 29 is from the tracking area 33 to the XY stage 105. If it jumps out in the x direction corresponding to the opposite direction of the moving direction, the beam is not irradiated to the end portion 29a on the −x direction side of the grid 29, so that the inspection cannot be performed. On the other hand, FIG. 15D scans the beam along the x direction. In this case, the grid 29 moves from the tracking area 33 to the XY stage 105, especially in scanning at the end 29b on the x-direction side of the grid 29, such as (19), (21) and (23). If it protrudes in the x direction corresponding to the opposite direction, the beam is not irradiated to the end portion 29b on the x direction side of the grid 29, so that the inspection cannot be performed. In order to avoid this problem, it is preferable to perform a beam scan in the y direction or the −y direction in the fourth step.

From the above, it is preferable that the beam scan along the x direction or the −x direction is performed in the second step or the third step. Then, the beam scan along the y direction or the −y direction is preferably performed in the first step or the fourth step.

FIG. 16 is a schematic diagram showing an example of a scanning operation in another comparative example of the first embodiment. In FIG. 16A, the beam is scanned along the y direction in the first step as shown in FIG. 11A, for example. However, unlike FIG. 11A, for example, scanning is performed with the end 29b on the x-direction side of the grid 29 as the start point and the end 29a on the −x direction side of the grid 29 as the end point. In this case, the grid 29 moves from the tracking area 33 to the XY stage 105, especially in scanning at the end 29b of the grid 29 on the x-direction side as shown in (1) and (2) of FIG. 16 (a). If it protrudes in the −x direction corresponding to the direction, the beam is not irradiated to the end portion 29b on the x-direction side of the grid 29, so that the inspection cannot be performed.

In FIG. 16 (b), the beam is scanned along the −y direction in the fourth step as shown in FIG. 11 (d), for example. However, unlike FIG. 11D, for example, scanning is performed with the end 29b on the x-direction side of the grid 29 as the start point and the end 29a on the −x direction side of the grid 29 as the end point. In this case, especially in scanning at the end 29a of the grid 29 on the −x direction side as in (23) and (24) of FIG. 16B, the grid 29 is moved from the tracking area 33 to the XY stage 105. If it jumps out in the x direction corresponding to the opposite direction of the moving direction, the beam is not irradiated to the end portion 29a on the −x direction side of the grid 29, so that the inspection cannot be performed.

From the above, it is preferable that the beam scan in the first step is performed with the end 29a on the −x direction side of the grid 29 as the start point and the end 29b on the x direction side of the grid 29 as the end point. The grid 29 continuously moves in the −x direction even during the beam scan. Therefore, when the beam scan is performed from the −x direction to the x direction, the grid 29 has already moved to the −x direction side when scanning the end of the grid 29 on the x direction side. Therefore, at this time, there is less possibility that the grid 29 will pop out from the tracking area 33 due to unintended vibration applied to the XY stage 105 and the inspection cannot be performed.

Similarly, the beam scan in the fourth step is preferably performed with the end 29a on the −x direction side of the grid 29 as the start point and the end 29b on the x direction side of the grid 29 as the end point. The grid 29 continuously moves in the −x direction even during the beam scan. Therefore, when the beam scan is performed from the −x direction to the x direction, the grid 29 is still in the position on the x direction side when scanning the end of the grid 29 on the −x direction side. Therefore, at this time, there is less possibility that the grid 29 will pop out from the tracking area 33 due to unintended vibration applied to the XY stage 105 and the inspection cannot be performed.

FIG. 17 is a flowchart showing a part of the main steps of the inspection method according to the first embodiment. FIG. 17 shows the steps from the start to the end of scanning among the main steps of the inspection method. In FIG. 17, the inspection methods according to the first embodiment are a substrate transfer step (S102), an inspection position moving step (S104), a stage moving step (constant velocity moving start) (S106), and a sub-region scanning step (first). 1 step) (S108a), sub-region scanning step (second step) (S108b), sub-region scanning step (third step) (S108c), sub-region scanning step (fourth step) ( S108d), a determination step (S110), a tracking reset step (S112), a determination step (S114), and a stripe movement step (S116).

In the substrate transfer step (S102), the substrate 101 is conveyed into the inspection room 103 and placed on the XY stage 105 by using a transfer mechanism (not shown).

As the inspection position moving step (S104), the drive mechanism 142 under the control of the stage control circuit 114 moves the XY stage 105 so that the inspection position is within the irradiable position of the multi-beam 20. First, the XY stage 105 is moved so that the irradiation region 34 of the multi-beam 20 is located on the left end side of the stripe region 32 (for example, two sizes outside the irradiation region 34).

As a stage moving step (start of constant velocity continuous movement) (S106), under the control of the stage control circuit 114, the drive mechanism 142 moves the XY stage 105 in the −x direction at a constant velocity V, for example. As a result, constant velocity continuous movement is started.

In the case of the embodiment shown in FIGS. 11 to 14, the sub-region scanning step (first step) (S108a), the sub-region scanning step (second step) (S108b), and the sub-region scanning step (third step). In (S108c) and the sub-region scanning step (fourth step) (S108d), in the electro-optical image acquisition mechanism 150a, the stripe region 32, which is the inspection region of the substrate 101, is p / M in the x direction (M is 2 or more). A plurality of grids 29 (sub-regions; small regions) having a size obtained by (integer) and divided by p (predetermined size) in the y direction are scanned for each N × N'grid 29 group. Specifically, among a plurality of grids 29, N × N'multi-beams are arranged in a group of N × N'grids on a substrate 101 in which N elements are arranged at a pitch p in the x direction and N'is arranged in the y direction. 20 are collectively deflected to start tracking, and the multi-beam 20 follows the continuous movement of the XY stage 105 while the XY stage 105 continuously moves the distance obtained by N / M · p in the −x direction. The N × N'grids 29 groups are scanned while tracking deflection. Then, the first secondary electronic image is acquired in the first step, the second secondary electronic image is acquired in the second step, and the third secondary electronic image is acquired in the third step. The fourth secondary electronic image is acquired in the fourth step. Further, the average secondary electronic image is acquired from the first secondary electronic image, the second secondary electronic image, the third secondary electronic image, and the fourth secondary electronic image by using the average image acquisition circuit 144. do. The acquired average secondary electronic image is stored in, for example, an image storage device 132. In the case of the embodiment shown in FIGS. 9 and 10, the sub-region scanning step (first step) (S108a) and the sub-region scanning step (second step) (S108b) are performed. Then, the first secondary electronic image is acquired in the first step, and the second secondary electronic image is acquired in the second step. Then, the average secondary electronic image is acquired from the first secondary electronic image and the second secondary electronic image by using the average image acquisition circuit 144.

FIG. 18 is a conceptual diagram for explaining another example of the details of the scanning operation according to the first embodiment. The multi-beam 20 has the same pitch p on the surface of the substrate 101 in the x direction (first direction), N rows (N is an integer of 2 or more), and the same pitch in the y direction (second direction) orthogonal to the x direction. It is composed of a plurality of electron beams arranged in an N'column (N'is an integer of 1 or more) in p (predetermined size). In the example of FIG. 18, as the N × N ′ multi-beams 20, 5 × 5 multi-beams arranged in the x-direction and the y-direction at the same pitch p on the surface of the substrate 101 are shown. The pitch in the y direction may be different from that in the x direction. One of the four adjacent measurement pixels 28 is a measurement pixel 28 as one of the four corners of the rectangle, and the p × p region surrounded by p × p in the x and y directions starting from the measurement pixel 28. 27 is divided in the x direction by the number of divisions M. Therefore, one grid 29 (sub-region; small region) is composed of a rectangular region having a size of p / M in the x direction and p in the −y direction. Therefore, in the example of FIG. 18, the case where the number of divisions M = 3 is shown. In the example of FIG. 18, each beam of 5 × 5 multi-beams scans the corresponding grid 29 (sub-region), and after scanning, N (5 in this case) are separated in the x direction. The case where the next grid 29 is scanned is shown.

First, the first function of the deflector 208, which is the basis of control by the deflection control circuit 128, is that the stripe region 32 (inspection region) of the substrate 101 is p / M (M is 2 or more) in the x direction by using the multi-beam 20. Of the plurality of grids 29 divided by the size obtained by (integer) and the size of p in the y direction, N × N'on the substrate 101 in which N pieces are arranged at pitch p in the x direction and N'pieces are lined up in the y direction. The multi-beam 20 is collectively deflected to the grid 29 group of the above. Here, among the plurality of grids 29 in the irradiation region 34 of the multi-beam 20, N × N ′ grids 29 groups arranged at a pitch p in the x direction are deflected as the tracking region 33. The main deflector 208 collectively deflects the multi-beam 20 to a reference position (for example, the center) of the tracking region 33. Then, the main deflector 208 tracks and deflects the multi-beam 20 so as to follow the continuous movement of the XY stage 105.

Under control by the deflection control circuit 128, the second function of the deflector 208 is that each beam of the multi-beam 20 is located on the corresponding grid 29, eg, the first pixel 36 in the x direction and the last pixel 36 in the y direction. The multi-beams 20 are collectively deflected so as to do so. Actually, the multi-beam 20 is collectively deflected so as to be located at the first pixel 36 in the x direction and the uppermost pixel 36 in the y direction in the scanning region 31 in which the margin is added to the corresponding grid 29. Then, while the multi-beam 20 is tracking-deflected so as to follow the continuous movement of the XY stage 105, N × N'grids 29 (specifically, the scanning region 31) set as the tracking region 33). The multi-beams 20 are collectively deflected to scan the group. Then, at each shot, each beam irradiates one measurement pixel 36 corresponding to the same position in the responsible grid 29. In the example of FIG. 18, each beam irradiates the first measurement pixel 36 from the left at the top of the grid 29 in charge of the first shot. Then, the deflector 208 shifts the beam deflection position of the entire multi-beam 20 in the −y direction by 36 minutes for one measurement pixel, and the second shot is 1 from the left in the second stage from the top in the grid 29 in charge. The third measurement pixel 36 is irradiated. After repeating such scanning and completing the irradiation of the first measurement pixel 36 from the left in the lowermost stage, the entire multi-beam 20 is collectively shifted in the x direction by one measurement pixel 36 minutes by the deflector 208 in the uppermost stage. The beam deflection position is shifted to the second measurement pixel 36 from the left. By repeating this operation, while the XY stage 105 continuously moves the distance obtained by N / M · p in the −x direction, one beam irradiates all the measurement pixels 36 in one grid 29 in order. go. Such an operation is simultaneously performed by N × N ′ multi-beams 20. In one shot, the multi-beam 20 formed by passing through each hole 22 of the molded aperture array substrate 203 allows the multi-beam 20 to be a bundle of secondary electrons corresponding to a plurality of shots having the same number as each hole 22 at the maximum. The secondary electron beam 300 is detected all at once. The first function of the deflector 208 is to move the XY stage 105 so that the movement of the XY stage 105 does not shift the deflection position until all the measurement pixels 36 in the grid 29 in charge of the multi-beam 20 are scanned. The multi-beam 20 is deflected (tracking operation) so as to follow.

As a detection step, the multi-detector 222 detects secondary electrons emitted from the substrate 101 due to irradiating the substrate 101 with the multi-beam 20. Each beam will scan one corresponding grid 29. Each shot of the multi-beam 20 causes secondary electrons to be emitted upward from the irradiated measurement pixel 36. In this way, the multi-detector 222 detects the secondary electrons emitted from the substrate 101 due to the irradiation of the substrate 101 with the multi-beam 20. The multi-detector 222 detects the multi-secondary electron beam 300 emitted upward from each measurement pixel 36 for each measurement pixel 36.

As a determination step (S110), when the multi-beam 20 scans all the measurement pixels 36 in the grid 29 (specifically, the scanning region 31) in charge of each of the multi-beams 20, the control computer 110 scans all the measurement pixels 36 in the target stripe region 32. It is determined whether or not the scanning of all the grids 29 is completed. When the scanning of all the grids 29 in the target stripe region 32 is completed, the process proceeds to the determination step (S114). If the scanning of all the grids 29 in the target stripe region 32 has not been completed, the process proceeds to the tracking reset step (S112).

As a tracking reset step (S112), by the deflection operation which is the second function of the deflector 208, all the measurement pixels 36 in the grid 29 (specifically, the scanning region 31) in charge of each are moved by the multi-beam 20. When scanned, the deflector 208 uses the first function to complete N × N'grids by the time the XY stages 105 have been moved in the −x direction at a distance of N / M · p. The multi-beam 20 is collectively re-biased to a new N × N'grid 29 (specifically, scanning region 31) group arranged at a pitch p in the x direction, which is N apart from the 29 group in the x direction. Perform tracking reset with.

In the example of FIG. 18, new N × N ′ grid 29 (specifically, scanning area 31) groups separated by 5 in the x direction are reset to the new tracking area 33. Then, the deflector 208 uses the first function to track and deflect the multi-beam 20 so as to follow the continuous movement of the XY stage 105. Further, at the time of tracking reset, the deflector 208 uses the second function, and each beam of the multi-beam 20 is the first and y in the corresponding grid 29 (specifically, the scanning region 31), for example, in the x direction. The multi-beam 20 is collectively deflected so that it is located at the final pixel 36 in the direction.

Then, the tracking cycle from the start of tracking to the reset of tracking and the scanning during tracking are repeated. In other words, as described above, N × N'grids 29 (specifically) set as the tracking region 33 while the multi-beam 20 is tracking deflected to follow the continuous movement of the XY stage 105. Scans the scanning area 31) group. By repeating this operation, all the pixels 36 in the stripe region 32 can be scanned.

Here, the first shot, the second shot, and the like are described separately, but the multi-beam 20 moves the deflection position while continuing to irradiate the beam for each pixel 36 without turning it on and off. A raster scan operation may be performed. Further, the present invention is not limited to the case where each pixel sequence arranged in the y direction is scanned in the same direction. The first row of pixels in the x direction may be scanned in the −y direction, and then the second row of pixels may be scanned in the y direction (opposite direction).

FIG. 19 is a conceptual diagram for explaining an example of the relationship between the sub-region and the corresponding beam in the scanning operation according to the first embodiment. In the example of FIG. 19, a case where 5 × 5 multi-beams 20 arranged at the same pitch p in the x and y directions are used is shown. Further, the case where the number of divisions M = 3 is shown. Further, in the example of FIG. 19, the width (y-direction size) of the stripe region 32 is divided according to the size of the irradiation region 34. Therefore, the stripe region 32 is divided into a plurality of grids 29 having a size in the x direction of p / M (= p / 3) and a size in the y direction of p. In the example of FIG. 19, the plurality of grids 29 are divided into five stages in the y direction, and the plurality of grids 29 in the x direction of each stage are N (= 5) in the x direction in charge of the same stage. The multi-beam 20 will scan. As shown in FIG. 19, five beams (1 to 5) in each stage scan the grid 29 during the nth tracking cycle in 1- (n), 2- (n), 3- (, respectively. It is shown by n), 4- (n), and 5- (n). When resetting the nth tracking, the deflector 208 re-deflects the N grids 29 ahead (= 5), so that the grid 29 5 ahead becomes the grid 29 scanned in the n + 1th tracking cycle. The grids 29 scanned during the n + 1th tracking cycle are indicated by 1- (n + 1), 2- (n + 1), 3- (n + 1), 4- (n + 1), and 5- (n + 1), respectively. All grids 29 can be scanned by repeating the same operation.

For example, in the p × p region 27 of p × p including the grid 29 that the first beam of the uppermost stage scans during the nth tracking cycle, there are the remaining two grids 29 divided into three. The grid adjacent to the 1- (n) grid 29 in the x direction (right side) will be scanned by the third beam during the n-1th tracking cycle. Further, the grid adjacent to the x direction (right side) will be scanned by the fifth beam during the n-2nd tracking cycle. The three tracking cycles complete the scan within the p × p region 27. The same applies to each stage in the y direction.

For example, in the p × p region 27 of p × p including the grid 29 that the second beam in the uppermost stage scans during the nth tracking cycle, there are the remaining two grids 29 divided into three. The grid adjacent to the 2- (n) grid 29 in the x direction (right side) will be scanned by the fourth beam during the n-1th tracking cycle. Further, the grid adjacent to the x direction (right side) will be scanned by the first beam during the n + 1th tracking cycle. By such three tracking cycles, the scan in the p × p region 27 of the p × p is completed. The same applies to each stage in the y direction.

For example, in the p × p region 27 of p × p including the grid 29 that the third beam in the uppermost stage scans during the nth tracking cycle, there are the remaining two grids 29 divided into three. The grid adjacent to the 3- (n) grid 29 in the x direction (right side) will be scanned by the fifth beam during the n-1th tracking cycle. Further, the grid adjacent to the x direction (right side) will be scanned by the second beam during the n + 1th tracking cycle. By such three tracking cycles, the scan in the p × p region 27 of the p × p is completed. The same applies to each stage in the y direction.

For example, in the p × p region 27 of p × p including the grid 29 that the fourth beam in the uppermost stage scans during the nth tracking cycle, there are the remaining two grids 29 divided into three. The grid adjacent to the 4- (n) grid 29 in the x direction (right side) will be scanned by the first beam during the n + second tracking cycle. Further, the grid adjacent to the x direction (right side) will be scanned by the third beam during the n + 1th tracking cycle. By such three tracking cycles, the scan in the p × p region 27 of the p × p is completed. The same applies to each stage in the y direction.

For example, in the p × p region 27 of p × p including the grid 29 that the fifth beam in the uppermost stage scans during the nth tracking cycle, there are the remaining two grids 29 divided into three. The grid adjacent to the 5- (n) grid 29 in the x direction (right side) will be scanned by the second beam during the n + second tracking cycle. Further, the grid adjacent to the x direction (right side) will be scanned by the fourth beam during the n + 1th tracking cycle. By such three tracking cycles, the scan in the p × p region 27 of the p × p is completed. The same applies to each stage in the y direction.

Therefore, when the stripe region 32 is scanned using the multi-beams of N = 5 with the number of divisions M = 3, the fifth beam outside the scanning start side end of the stripe region 32 is at least the tracking cycle 2. The scanning operation is started from a position where the grid 29 on the front side of the batch can be scanned.

By scanning using the multi-beam 20 as described above, a scanning operation (measurement) can be performed at a higher speed than when scanning with a single beam.

As the determination step (S114), the control computer 110 determines whether or not the scanning of all the stripes 32 is completed. When the scanning of all the stripes 32 is completed, the electro-optical image acquisition process is completed. If the scanning of all the stripes 32 is not completed, the process proceeds to the stripe moving step (S116).

In the stripe moving step (S116), the drive mechanism 142, which is the control base of the stage control circuit 114, is located on the left end side of the next stripe region 32 (for example, outside by one size of the irradiation region 34) of the irradiation region of the multi-beam 20. The XY stage 105 is moved so that 34 is located. Then, each step described above is repeated.

FIG. 20 is a conceptual diagram for explaining another example of the scanning operation according to the first embodiment. As shown in FIG. 20, in the inspection region 330 of the substrate 101, for example, a plurality of chips 332 (dies) having a predetermined width are formed in an array in the x and y directions. Here, it is preferable that the substrate 101 to be inspected is adapted to a semiconductor substrate (for example, a wafer). Each chip 332 is formed on the substrate 101 in a size of, for example, 30 mm × 25 mm. The pattern inspection will be performed for each chip 332. The region of each chip 332 is virtually divided into a plurality of stripe regions 32 having the same width in the y direction as the irradiation region 34 that can be irradiated by one irradiation of the entire multi-beam 20. The scanning operation of each stripe area 32 may be the same as the above-described content. As described above, in the first embodiment, each stripe region 32 is scanned by the multi-beam 20 while the irradiation region 34 is relatively continuously moved in the x direction by continuously moving the XY stage 105 in the −x direction. go. When the scanning of all the stripe regions 32 is completed, the stage position is moved in the −y direction, and the stripe regions 32 of the same next stage in the y direction are similarly scanned by the multi-beam 20. When the scanning of the region of one chip 332 is completed by repeating this operation, the XY stage 105 is moved to perform the scanning operation on the stripe region 32 at the upper end of the next chip 332 in the same manner. By repeating this operation, all the chips 332 are scanned.

As described above, the electro-optical image acquisition mechanism 150a scans the substrate 101 on which the graphic pattern is formed by using the multi-beam 20 with a plurality of electron beams while continuously moving the XY stage 105, and the multi-beam 20 is generated. The multi-secondary electron beam 300 emitted from the substrate 101 due to the irradiation is detected. The method of scanning and the method of detecting the multi-secondary electron beam 300 are as described above. The detection data of secondary electrons from each measurement pixel 36 detected by the multi-detector 222 is output to the detection circuit 106 in the order of measurement. In the detection circuit 106, analog detection data is converted into digital data by an A / D converter (not shown) and stored in the stripe pattern memory 123. Then, at the stage where the detection data of one stripe region 32 minutes (or chip 332 minutes) is accumulated, the comparison circuit 108 is used as stripe pattern data (or chip pattern data) together with information indicating each position from the position circuit 107. Transferred to.

On the other hand, a reference image is created in parallel with or before and after the multi-beam scan and secondary electron detection steps.

As a reference image creation step, when the substrate 101 is a semiconductor substrate, the reference image creation unit such as the development circuit 111 and the reference circuit 112 exposes an exposure image on the substrate when the mask pattern of the exposure mask is exposed and transferred to the semiconductor substrate. Based on the exposure image data in which is defined, a reference image of a region corresponding to a measurement image (electro-optical image) of a frame region described later having a size of a grid 29 or less composed of a plurality of pixels 36 is created. Instead of the exposure image data, drawing data (design data) that is a basis for forming an exposure mask that exposes and transfers a plurality of graphic patterns to the substrate 101 may be used. When the substrate 101 is an exposure mask, the reference image creating unit such as the expansion circuit 111 and the reference circuit 112 is based on drawing data (design data) that is a source for forming a plurality of graphic patterns on the substrate 101. A reference image of a region corresponding to a measurement image (electro-optical image) of a frame region composed of a plurality of pixels 36 is created.

Specifically, it operates as follows. First, the expansion circuit 111 reads drawing data (or exposure image data) from the storage device 109 through the control computer 110, and outputs each graphic pattern of each frame area defined in the read drawing data (or exposure image data). It is converted into binary or multi-valued image data, and this image data is sent to the reference circuit 112.

Here, the figure defined in the drawing data (or exposure image data) is, for example, a rectangle or a triangle as a basic figure, for example, the coordinates (x, y), the side length, and the rectangle at the reference position of the figure. Graphic data that defines the shape, size, position, etc. of each pattern graphic is stored with information such as a graphic code that serves as an identifier that distinguishes graphic types such as triangles and triangles.

When the drawing data (or exposure image data) that becomes the graphic data is input to the expansion circuit 111, it is expanded to the data for each graphic, and the graphic code, the graphic dimension, etc. indicating the graphic shape of the graphic data are interpreted. Then, binary or multi-valued design image data is developed and output as a pattern arranged in a grid having a grid of predetermined quantization dimensions as a unit. In other words, the design data is read, the inspection area is virtually divided into squares with a predetermined dimension as a unit, the occupancy rate of the figure in the design pattern is calculated for each square, and the n-bit occupancy rate data is obtained. Output. For example, it is preferable to set one square as one pixel. Then, when to have a resolution of 1/2 8 ^{(= 1/256)} to 1 pixel, the occupancy rate of the pixel allocated the small area region amount corresponding 1/256 of figures are arranged in a pixel Calculate. Then, it is output to the reference circuit 112 as 8-bit occupancy rate data. Such squares may have the same size as the measurement pixel 36.

Next, the reference circuit 112 applies appropriate filtering to the design image data, which is the image data of the sent graphic figure. Since the measurement data as an optical image obtained from the detection circuit 106 is in a state in which the filter is acted by the electro-optical system, in other words, in an analog state in which the image intensity (shade value) changes continuously, the image on the design side where the image intensity (shade value) is a digital value. By filtering the design image data, which is the data, it can be matched with the measurement data. In this way, a design image (reference image) to be compared with the measurement image (optical image) of the frame region is created. The image data of the created reference image is output to the comparison circuit 108, and the reference images output in the comparison circuit 108 are stored in the memory, respectively.

FIG. 21 is a diagram showing an internal configuration of the comparison circuit according to the first embodiment. In FIG. 21, storage devices 50 and 52 such as a magnetic disk device, a division unit 56, an alignment unit 58, and a comparison unit 60 are arranged in the comparison circuit 108. Each "-part" such as the division part 56, the alignment part 58, and the comparison part 60 includes a processing circuit, and the processing circuit includes an electric circuit, a computer, a processor, a circuit board, a quantum circuit, a semiconductor device, or the like. Is included. Further, a common processing circuit (same processing circuit) may be used for each “~ part”. Alternatively, different processing circuits (separate processing circuits) may be used. The input data or the calculated result required in the division unit 56, the alignment unit 58, and the comparison unit 60 are stored in a memory (not shown) each time.

The transferred stripe pattern data (or chip pattern data) is temporarily stored in the storage device 50 together with information indicating each position from the position circuit 107. Similarly, the reference image data is temporarily stored in the storage device 52 together with information indicating each position in the design.

Next, the division unit 56 divides the stripe pattern data (or chip pattern data) into each frame area (unit inspection area) to generate a plurality of frame images (inspection images).

FIG. 22 is a diagram showing an example of the relationship between the grid and the frame area in the first embodiment. As mentioned above, each grid 29 is scanned by one beam. As described above, each grid 29 is scanned for each scanning region 31 that is larger than the grid 29 so that there is no joint error with the grid 29 scanned by another beam. Since the characteristics of the obtained secondary electron image may shift from beam to beam, it is desirable to use the image obtained by one beam as the image of one unit inspection area. Therefore, in the first embodiment, the measured stripe pattern data (or chip pattern data) is divided for each frame area 35 which is a unit inspection area, and a plurality of frame images are created. Similarly, a reference image is created for each frame area 35. In that case, the frame area 35 is set to a range scanned by one beam. Therefore, the frame area 35 is set to the grid 29 size or less. For example, it may be set to a size that is a natural number of the grid 29. In this way, the dividing unit 56 divides the detected secondary electron image into inspection images having a size equal to or smaller than that of the scanning region 31.

Next, the alignment unit 58 aligns the frame image (measured image) and the reference image in units of sub-pixels smaller than the pixel 36. For example, the alignment may be performed by the method of least squares.

Then, the comparison unit 60 compares the frame image (inspection image) with the reference image corresponding to the frame image (average secondary electron image). For example, the comparison unit 60 compares the frame image and the reference image for each pixel 36. The comparison unit 60 compares the two for each pixel 36 according to a predetermined determination condition, and determines the presence or absence of a defect such as a shape defect. For example, if the gradation value difference for each pixel 36 is larger than the determination threshold value, it is determined as a defect. Alternatively, the inspection accuracy may be lower than that of the shape defect inspection, and the presence or absence of disconnection or short circuit of the pattern may be inspected. For example, the edge pair of the pattern is detected and the distance between the edge pairs is measured. Thereby, the width dimension of the line pattern and the distance of the space portion between the line patterns can be measured. Similarly, if the difference from the distance obtained from the reference image is larger than the determination threshold value, it is determined to be a defect. By measuring between a plurality of edge pairs in the line direction, it is possible to inspect the presence or absence of disconnection and / or short circuit of the pattern. Then, the comparison result is output. The comparison result may be output from the storage device 109, the monitor 117, the memory 118, or the printer 119.

As described above, according to the first embodiment, the swing width of the beam deflection is reduced in the pattern inspection performed while the XY stage 105 is continuously moved by using the multi-beam 20 in which a plurality of beams are arranged in the moving direction of the XY stage 105. can. Therefore, the influence of the aberration of the optical system can be suppressed. Further, by using the multi-beam 20, the throughput can be improved in the pattern inspection.

FIG. 23 is a configuration diagram showing another configuration of the pattern inspection device according to the first embodiment. The electro-optical image acquisition mechanism 150a shown in FIG. 1 includes a deflector 208. On the other hand, the electro-optical image acquisition mechanism 150b shown in FIG. 23 includes a main deflector 208 and a sub-deflector 209. For example, the main deflector 208 is responsible for the first function, and the sub-deflector 209 is responsible for the second function.

For example, the inspection device according to the configuration of FIG. 23 includes a movable stage on which a substrate is placed, a stage control circuit that continuously moves the stage in the opposite direction of the first direction, and a substrate surface in the first direction. Is composed of a plurality of charged particle beams arranged in N rows (N is an integer of 2 or more) and N'rows (N'is an integer of 1 or more) in the second direction orthogonal to the first direction at the same pitch p. Of a plurality of small regions in which the inspection region of the substrate is divided in the first direction by p / M (M is an integer of 2 or more) and the second direction by a predetermined size by using the multi-beam. , The multi-beam is collectively deflected to a small region group of N × N'on a substrate in which N pieces are arranged at a pitch p in the first direction and N'pieces are lined up in the second direction, and the stage is set in the first direction. While continuously moving the distance obtained by N / M ・ p in the opposite direction of, the multi-beam is tracked and deflected to follow the continuous movement of the stage, and N / M ・ p is used in the opposite direction of the first direction. By the time the movement of the stage at the obtained distance is completed, N × N'small regions are separated from the N × N'small region group in the first direction by N, and new N × N'arranged at a pitch p in the first direction. A first deflector (main deflector 208) that resets tracking by collectively re-deflecting the multi-beams into a small region group, and while the multi-beams are tracking-deflected to follow the continuous movement of the stage. In addition, each of the multi-beams, in each of the plurality of subregions, starting from the end on the opposite side of the first direction in each of the plurality of subregions, and in each of the plurality of subregions. With the end on the side in the first direction as the end point, the collective deflection of the multi-beam along the second direction is applied from the end on the side opposite to the first direction to the end on the side in the first direction. The first step is repeated toward, and then the second step is repeated, in which the multi-beams are collectively deflected along the opposite direction of the first direction, and then along the first direction. A third step of repeatedly deflecting the multi-beams in a batch is performed, and then, in each of the plurality of small regions, starting from the end on the side opposite to the first direction and in the plurality of small regions. In each case, with the end on the side in the first direction as the end point, the collective deflection of the multi-beam along the opposite direction in the second direction is applied to the first from the end on the side in the opposite direction in the first direction. By performing a fourth step that is repeated towards the end on the directional side
A second deflector (secondary deflector 209) that collectively deflects the multi-beams so as to scan N × N'small region groups, and the substrate due to the irradiation of the multi-beams on the substrate. A detector for detecting the emitted secondary electrons is provided, and the value of the combination of the value of N and the value of M, in which the greatest common divisor between the value of N and the value of M is 1, is used. It is a characteristic inspection device.

In the above description, the series of "~ circuits" includes a processing circuit, and the processing circuit includes an electric circuit, a computer, a processor, a circuit board, a quantum circuit, a semiconductor device, and the like. Further, a common processing circuit (same processing circuit) may be used for each "~ circuit". Alternatively, different processing circuits (separate processing circuits) may be used. The program for executing the processor or the like may be recorded on a recording medium such as a magnetic disk device, a magnetic tape device, an FD, or a ROM (read-only memory). Further, the series of "storage devices" of the present embodiment are, for example, a magnetic disk device, a magnetic tape device, an FD, a ROM, a flash memory, and the like.

The embodiment has been described above with reference to a specific example. However, the present invention is not limited to these specific examples. In the above-mentioned example, the case where the XY stage 105 is continuously moved at a constant speed is shown, but the present invention is not limited to this. Constant velocity continuous movement is desirable from the viewpoint of ease of control, but continuous movement accompanied by acceleration / deceleration may be used. Further, the magnitude relationship between the number of beams N and the number of divisions M in the continuous moving direction (-x direction in the embodiment) of the XY stage 105 during scanning is such that the greatest common divisor between the two values is 1. It doesn't matter which one is bigger.

Further, the irradiation order of the beams when scanning in the grid 29 (scanning area 31) may be arbitrary. However, since the entire multi-beam 20 is deflected collectively by the deflector 208, the irradiation order is the same between the grids 29.

Further, in the above-mentioned example, the case where the beam arrangement is an orthogonal lattice is shown, but the present invention is not limited to this. For example, an oblique grid may be used. Alternatively, the beam trains arranged in the x direction may be arranged slightly offset in the x direction in each stage adjacent to each other in the y direction. For example, the heads of the beam trains arranged in the x direction may be arranged unevenly.

Further, the shape of the grid 29 (sub-region; small region) described above is not limited to the case of being rectangular. Other shapes may be used as long as they are arranged at a pitch of p / M in the x direction and at equal pitches in the y direction. The shape of the grid 29 (sub-region; small region) may be, for example, a parallelogram. In such a case, it is preferable to set the scanning area 31 to be a parallelogram according to the shape of the grid 29. For parallelograms. Since the circuit pattern has many horizontal or / or orthogonal lines in the x direction, it is easy to avoid scanning in parallel with the circuit pattern by scanning the beam diagonally, and the influence of charging can be avoided.

Further, the arrangement pitch of the multi-beam 20 may be different in the x direction and the y direction. For example, they may be arranged at equal pitch p in the x direction and at equal pitch p'in the y direction.

In addition, although the description of parts that are not directly necessary for the description of the present invention, such as the device configuration and control method, is omitted, the required device configuration and control method can be appropriately selected and used.

In addition, all electron beam inspection devices and electron beam inspection methods that include the elements of the present invention and can be appropriately redesigned by those skilled in the art are included in the scope of the present invention.

20 Multi-beam 22 holes 27 area (p × p area)
28, 36 Pixels 29 Grid 30,330 Inspection area 31 Scan area 32 Stripe area 33 Tracking area 34 Irradiation area 35 Frame area 50, 52 Storage device 56 Division 58 Alignment unit 60 Comparison unit 100 Inspection device 101 Board 102 Electron beam column 103 Inspection room 106 Detection circuit 107 Position circuit 108 Comparison circuit 109 Storage device 110 Control computer 111 Deployment circuit 112 Reference circuit 114 Stage control circuit 117 Monitor 118 Memory 119 Printer 122 Laser length measurement system 120 Bus 123 Stripe pattern memory 124 Lens control circuit 126 Blanking control circuit 128 Deflection control circuit 132 Image storage device 140 Settling time storage device 141 Scan frequency storage device 144 Average image acquisition circuit 146 Movement speed calculation circuit 148 Scan time calculation circuit 149 Scan time storage device 150a Electro-optical image acquisition mechanism 150b Electronic Optical image acquisition mechanism 160 Control system circuit 200 Electron beam 201 Electron gun 202 Electron lens 203 Molded aperture array substrate 205 Electromagnetic lens (reduction lens)
206 Electromagnetic lens 207 Electromagnetic lens (objective lens)
208 deflector (main deflector)
209 Sub-deflector 212 Collective blanking deflector 214 Beam separator 216 Mirror 222 Multi-detector 224,226 Projection lens 228 Deflection 300 Multi-secondary electron 332 chip

Claims (20)

A movable stage on which the board is placed,
A stage control circuit that continuously moves the stage in the opposite direction of the first direction,
N rows (N is an integer of 2 or more) at the same pitch p on the substrate surface in the first direction and N'rows (N'is an integer of 1 or more) in the second direction orthogonal to the first direction. ) Using a multi-beam composed of a plurality of charged particle beams arranged side by side, the inspection region of the substrate has a size obtained in p / M (M is an integer of 2 or more) in the first direction and the second direction. Of the plurality of small regions divided into predetermined sizes, N small regions on the substrate in which N in the first direction at the pitch p and N'in the second direction are arranged. The multi-beams are collectively deflected into groups so that the stage follows the continuous movement of the stage while continuously moving the distance obtained at N / M · p in the opposite direction of the first direction. From the N × N'small region group by the time the multi-beam is track-deflected and the movement of the stage at the distance obtained by the N / M · p is completed in the direction opposite to the first direction. Tracking reset is performed by collectively re-deflecting the multi-beams into a new group of N × N'small regions arranged in the first direction at the pitch p, which are N apart from each other in the first direction. The first function and
While the multi-beam is tracking deflected to follow the continuous movement of the stage, each of the multi-beams is placed in each of the plurality of small regions.
The start point is the end on the side opposite to the first direction in each of the plurality of small regions, and the end on the side in the first direction in each of the plurality of small regions is the end point. , The first batch of deflection of the multi-beam along the second direction is repeated from the end on the side opposite to the first direction toward the end on the side in the first direction. Perform the process, then
The start point is the end on the side opposite to the first direction in each of the plurality of small regions, and the end on the side in the first direction in each of the plurality of small regions is the end point. , The collective deflection of the multi-beam along the opposite direction of the second direction is repeatedly performed from the end on the side opposite to the first direction toward the end on the side in the first direction. By performing the second step
A deflector having two functions of a second function of collectively deflecting the multi-beam so as to scan the N × N'small region group,
A detector that detects secondary electrons emitted from the substrate due to irradiation of the substrate with the multi-beam, and a detector.
With
A charged particle beam inspection apparatus, wherein as the value of N and the value of M, a combination value in which the greatest common divisor between the value of N and the value of M is 1 is used. The measurement pixel size that can be irradiated by the charged particle beam is PS, the second direction or the second direction including the second beam setting time Ofs_v in the second direction or the opposite direction of the second direction. When the beam scan time in the opposite direction is Tv,
The moving speed V of the stage is V = PS (M-1) / (2Tv).
The charged particle beam inspection apparatus according to claim 1. When the measurement pixel size that can be irradiated with the charged particle beam is PS and the scan frequency of the charged particle beam is f,
The beam scan time Tv in the second direction or the opposite direction of the second direction is Tv = (p / PS) × (p / M / PS) × (1 / f) + (p / M / PS) × Ofs_v
The charged particle beam inspection apparatus according to claim 1. An average image acquisition for acquiring an average secondary electron image obtained by averaging the first secondary electron image acquired in the first step and the second secondary electron image acquired in the second step. The charged particle beam inspection apparatus according to claim 1, further comprising a circuit. A movable stage on which the board is placed,
A stage control circuit that continuously moves the stage in the opposite direction of the first direction,
N rows (N is an integer of 2 or more) at the same pitch p on the substrate surface in the first direction and N'rows (N'is an integer of 1 or more) in the second direction orthogonal to the first direction. ) Using a multi-beam composed of a plurality of charged particle beams arranged side by side, the inspection region of the substrate has a size obtained in p / M (M is an integer of 2 or more) in the first direction and the second direction. Of the plurality of small regions divided into predetermined sizes, N small regions on the substrate in which N in the first direction at the pitch p and N'in the second direction are arranged. The multi-beams are collectively deflected into groups so that the stage follows the continuous movement of the stage while continuously moving the distance obtained at N / M · p in the opposite direction of the first direction. From the N × N'small region group by the time the multi-beam is track-deflected and the movement of the stage at the distance obtained by the N / M · p is completed in the direction opposite to the first direction. Tracking reset is performed by collectively re-deflecting the multi-beams into a new group of N × N'small regions arranged in the first direction at the pitch p, which are N apart from each other in the first direction. The first function and
While the multi-beam is tracking deflected to follow the continuous movement of the stage, each of the multi-beams is placed in each of the plurality of small regions.
The start point is the end on the side opposite to the first direction in each of the plurality of small regions, and the end on the side in the first direction in each of the plurality of small regions is the end point. , The first batch of deflection of the multi-beam along the second direction is repeated from the end on the side opposite to the first direction toward the end on the side in the first direction. Perform the process, then
A second step of repeating the batch deflection of the multi-beam along the opposite direction of the first direction is performed, and then
A third step of repeating the batch deflection of the multi-beam along the first direction is performed, and then
The start point is the end on the side opposite to the first direction in each of the plurality of small regions, and the end on the side in the first direction in each of the plurality of small regions is the end point. , The collective deflection of the multi-beam along the opposite direction of the second direction is repeatedly performed from the end on the side opposite to the first direction toward the end on the side in the first direction. By performing the fourth step
A deflector having two functions of a second function of collectively deflecting the multi-beam so as to scan the N × N'small region group,
A detector that detects secondary electrons emitted from the substrate due to irradiation of the substrate with the multi-beam, and a detector.
With
A charged particle beam inspection apparatus, wherein as the value of N and the value of M, a combination value in which the greatest common divisor between the value of N and the value of M is 1 is used. The charged particle beam inspection according to claim 5, wherein the collective deflection of the multi-beam in the second step is repeated from the side opposite to the second direction to the side in the second direction. Device. The charged particle beam inspection according to claim 5, wherein the collective deflection of the multi-beam in the second step is repeated from the side in the second direction to the side in the opposite direction of the second direction. Device. The measurement pixel size that can be irradiated with the charged particle beam is PS,
Th.
When the beam scan time in the second direction or the reverse direction of the second direction including the second beam setting time Ofs_v in the second direction or the reverse direction of the second direction is Tv,
The moving speed V of the stage is V = PS (M-1) / (2 × (M × Th + Tv)).
The charged particle beam inspection apparatus according to claim 5. When the measurement pixel size that can be irradiated with the charged particle beam is PS and the scan frequency of the charged particle beam is f,
The beam scan time Th in the first direction or the opposite direction of the first direction is Th = (p / PS) × (p / M / PS) × (1 / f) + (p / PS) × Ofs_h.
And
The beam scan time Tv in the second direction or the opposite direction of the second direction is Tv = (p / PS) × (p / M / PS) × (1 / f) + (p / M / PS) × Ofs_v
The charged particle beam inspection apparatus according to claim 5. The first secondary electronic image acquired in the first step, the second secondary electronic image acquired in the second step, and the third secondary electronic image acquired in the third step. The charged particle beam inspection apparatus according to claim 5, further comprising an average image acquisition circuit for acquiring an average secondary electron image obtained by averaging the electron image and the fourth secondary electron image acquired in the fourth step. .. N rows (N is an integer of 2 or more) and N'rows (N'is an integer of 1 or more) are arranged in the first direction on the substrate surface at the same pitch p in the second direction orthogonal to the first direction. Using a multi-beam composed of a plurality of charged particle beams, the inspection region of the substrate has a size obtained in p / M (M is an integer of 2 or more) in the first direction and is predetermined in the second direction. Among a plurality of small regions divided by size, N × N'small regions on the substrate are arranged in N in the first direction at the pitch p and N'in the second direction. The multi-beams are collectively deflected to follow the continuous movement of the stage while the stage on which the substrate is placed continuously moves the distance obtained by N / M · p in the opposite direction of the first direction. While tracking and deflecting the multi-beam as
Each of the multi-beams in each of the plurality of small regions
The start point is the end on the side opposite to the first direction in each of the plurality of small regions, and the end on the side in the first direction in each of the plurality of small regions is the end point. , The first batch of deflection of the multi-beam along the second direction is repeated from the end on the side opposite to the first direction toward the end on the side in the first direction. Perform the process, then
The start point is the end on the side opposite to the first direction in each of the plurality of small regions, and the end on the side in the first direction in each of the plurality of small regions is the end point. , The collective deflection of the multi-beam along the opposite direction of the second direction is repeatedly performed from the end on the side opposite to the first direction toward the end on the side in the first direction. By performing the second step
Secondary electrons emitted from the substrate due to irradiating the substrate with the multi-beam are detected.
By the time the movement of the stage at the distance obtained by N / M · p in the opposite direction of the first direction was completed, N pieces were separated from the N × N'small region group in the first direction. A charged particle beam inspection method in which tracking reset is performed by collectively re-deflating the multi-beam into a new N × N'small region group arranged in the first direction at the pitch p.
A charged particle beam inspection method using as the value of N and the value of M a combination of values in which the greatest common divisor between the value of N and the value of M is 1. The measurement pixel size that can be irradiated by the charged particle beam is PS, the second direction or the second direction including the second beam setting time Ofs_v in the second direction or the opposite direction of the second direction. When the beam scan time in the opposite direction is Tv,
The moving speed V of the stage is V = PS (M-1) / (2Tv).
The charged particle beam inspection method according to claim 11. When the measurement pixel size that can be irradiated with the charged particle beam is PS and the scan frequency of the charged particle beam is f,
The beam scan time Tv in the second direction or the opposite direction of the second direction is Tv = (p / PS) × (p / M / PS) × (1 / f) + (p / M / PS) × Ofs_v
The charged particle beam inspection method according to claim 11. 11. Claim 11 to acquire an average secondary electronic image obtained by averaging the first secondary electronic image acquired in the first step and the second secondary electronic image acquired in the second step. The charged particle beam inspection device described. N rows (N is an integer of 2 or more) and N'rows (N'is an integer of 1 or more) are arranged in the first direction on the substrate surface at the same pitch p in the second direction orthogonal to the first direction. Using a multi-beam composed of a plurality of charged particle beams, the inspection region of the substrate has a size obtained in p / M (M is an integer of 2 or more) in the first direction and is predetermined in the second direction. Among a plurality of small regions divided by size, N × N'small regions on the substrate are arranged in N in the first direction at the pitch p and N'in the second direction. The multi-beams are collectively deflected to follow the continuous movement of the stage while the stage on which the substrate is placed continuously moves the distance obtained by N / M · p in the opposite direction of the first direction. While tracking and deflecting the multi-beam as
Each of the multi-beams in each of the plurality of small regions
The start point is the end on the side opposite to the first direction in each of the plurality of small regions, and the end on the side in the first direction in each of the plurality of small regions is the end point. , The first batch of deflection of the multi-beam along the second direction is repeated from the end on the side opposite to the first direction toward the end on the side in the first direction. Perform the process, then
A second step of repeating the batch deflection of the multi-beam along the opposite direction of the first direction is performed, and then
A third step of repeating the batch deflection of the multi-beam along the opposite direction of the first direction is performed, and then
The start point is the end on the side opposite to the first direction in each of the plurality of small regions, and the end on the side in the first direction in each of the plurality of small regions is the end point. , The collective deflection of the multi-beam along the opposite direction of the second direction is repeatedly performed from the end on the side opposite to the first direction toward the end on the side in the first direction. By performing the fourth step
Secondary electrons emitted from the substrate due to irradiating the substrate with the multi-beam are detected.
By the time the movement of the stage at the distance obtained by N / M · p in the opposite direction of the first direction was completed, N pieces were separated from the N × N'small region group in the first direction. A charged particle beam inspection method in which tracking reset is performed by collectively re-deflating the multi-beam into a new N × N'small region group arranged in the first direction at the pitch p.
A charged particle beam inspection method using as the value of N and the value of M a combination of values in which the greatest common divisor between the value of N and the value of M is 1. The charged particle beam inspection according to claim 15, wherein the collective deflection of the multi-beam in the second step is repeated from the side opposite to the second direction to the side in the second direction. Method. The charged particle beam inspection according to claim 15, wherein the collective deflection of the multi-beam in the second step is repeated from the side in the second direction to the side in the opposite direction of the second direction. Method. The measurement pixel size that can be irradiated with the charged particle beam is PS,
Th.
When the beam scan time in the second direction or the reverse direction of the second direction including the second beam setting time Ofs_v in the second direction or the reverse direction of the second direction is Tv,
The moving speed V of the stage is V = PS (M-1) / (2 × (M × Th + Tv)).
The charged particle beam inspection method according to claim 15. When the measurement pixel size that can be irradiated with the charged particle beam is PS and the scan frequency of the charged particle beam is f,
The beam scan time Th in the first direction or the opposite direction of the first direction is Th = (p / PS) × (p / M / PS) × (1 / f) + (p / PS) × Ofs_h.
And
The beam scan time Tv in the second direction or the opposite direction of the second direction is Tv = (p / PS) × (p / M / PS) × (1 / f) + (p / M / PS) × Ofs_v
The charged particle beam inspection method according to claim 15. The first secondary electron image acquired in the first step, the second secondary electron image acquired in the second step, and the third secondary electron image acquired in the third step. The charged particle beam inspection method according to claim 15, wherein an average secondary electron image obtained by averaging the electron image and the fourth secondary electron image acquired in the fourth step is obtained.

Images