JP2003031629A - Pattern inspection method and apparatus - Google Patents

Abstract

[PROBLEMS] To irradiate a target particle substrate with a charged particle beam, detect secondary electrons and the like to generate an image, obtain an image, and compare the obtained image with a reference image to remove a pattern defect. In the method of inspecting or moving to a pattern defect position and confirming a defect, the target substrate is charged up by charged particle irradiation, the charged particle beam is bent, and the defect position shifts. A defect position at the time of inspection is corrected by calculating a position deviation amount at the time of inspection and a position deviation amount from a reference image at the same time as the inspection. Further, an image at the time of inspection is added to the defect information, and the defect information is corrected based on a positional shift amount between the image information and the design information. Further, an image at the time of inspection is added to the defect information, and the position is corrected by collating the image with the image at the time of defect confirmation. Further, an image is acquired at the time of defect confirmation, added to the defect information, and the defect position is corrected by comparing the image at the time of defect confirmation with an image detected again by another device or design information.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a substrate manufacturing apparatus having a circuit pattern such as a semiconductor device or a liquid crystal, and in particular, a pattern of a substrate being manufactured, that is, a defect on an insulating film or a conductor pattern isolated on the insulating film. Technology related to the inspection of defects.

[0002]

2. Description of the Related Art A conventional technique of an electron beam type pattern inspection apparatus is known, for example, in Japanese Unexamined Patent Publication No. 5-258703 (Prior Art 1). In this prior art 1, an electron beam from an electron beam source is deflected in a X direction by a deflector, irradiates a target substrate through an objective lens, and at the same time, a stage is continuously moved (traveled) in the Y direction, The secondary electrons from the substrate are detected by the detector, and the detection signal is detected by the A / D converter.
A digital image is subjected to D / D conversion, and compared with a reference digital image at a place where the image processing circuit can be expected to be originally the same, a difference is detected as a pattern defect, and the defect position is determined.

As a conventional technique of an optical inspection apparatus, for example, Japanese Patent Laid-Open No. 11-160247 (Prior Art 2) is known.
Known in. In this prior art 2, the light from the light source is applied to the target substrate through the objective lens, and the reflected light at that time is detected by the image sensor. An image is detected as a detected image by repeating detection while moving (running) the stage at a constant speed, and stored in the memory. The image is compared with a stored image on the memory that can be expected to have the same pattern as the detected image, and if the pattern is the same, a normal portion is detected, and if the pattern is different, it is detected as a pattern defect and the defect position is determined.

In Japanese Patent Laid-Open No. 9-320505 (Prior Art 3) and Japanese Patent Laid-Open No. 2000-208085 (Prior Art 4), the conditions of various electron optical systems are changed to minimize charge-up. The detection method is described.

Further, Japanese Patent Laid-Open No. 10-318950
In (Prior Art 5), the difference image level from the pixel of two images to be compared, the change rate of the local gradation value of the images, and the representative value of the local gradation values are used to calculate the difference image level. The permissible range for the tonal value is obtained for each pixel, and the gradation value of the difference image is compared with the permissible range calculated for each pixel. Pixels whose gradation value of the difference image is within the permissible range are non-defect candidates, permissible range. It is described that pixels outside are determined as defect candidates.

Further, Japanese Patent Laid-Open No. 11-251377 (Prior Art 6) discloses a design of an image at the time of inspection, a dictionary image registered in advance in order to correspond to a stage accuracy and a wafer pattern arrangement error, and a dictionary image position. It is described that the defect coordinates are corrected using the position information.

Machine Vision and Applications
(1989) PP155-166 `` Knowledge-Directed Inspection f
"or Complex Mutilayed Patterns" (prior art 7) describes the alignment of the detected image and the design information.

[0008]

In the inspection using an electron beam, an image obtained by irradiating a circuit board with an electron beam and imaging secondary electrons, backscattered electrons, transmitted electrons and absorption currents generated at that time Based on the above, it is determined whether or not there is a defect in the pattern of the circuit board.

By the way, when the circuit board is irradiated with an electron beam, if the secondary electron emission efficiency η is other than 1, the circuit board is charged positively or negatively. However, in the case where the circuit board is a conductor and the conductor escapes to the stage on which the circuit board is mounted, no electric charge is accumulated and the above-mentioned problem of charging does not occur.

However, when the insulating material or the conductor is isolated in the insulating material or the like, the charge is accumulated. Due to this accumulated charge, the electron beam to be irradiated next is deflected. When deflected, the position where you think that you are irradiating is different from the position where the actual image is detected,
There is a problem that the inspection itself is impossible or the detected defect position is different. In addition, when the defect is confirmed after the inspection is completed, there is a problem that the defect position is displaced due to charging and the defect cannot be confirmed.

Further, in the above-mentioned prior arts 3 and 4, the method of irradiating secondary electrons is devised to eliminate the charging itself. However, it is extremely difficult to completely eliminate the charging. That is, even if the condition for eliminating the electrostatic charge can be realized by the combination of a specific material and the pattern density, the circuit board to be inspected has a different pattern density or a different material depending on the location. It is impossible to dynamically change the conditions according to these conditions at the current technical level. In other words, it cannot be said that the conventional inspection method and apparatus have sufficiently taken into consideration these charging problems.

Further, in the method of the prior art 5 described above, the steady error is corrected by using the design information, but the design information is necessary at the time of inspection, and the defect confirmation is not described. It could not be said that the feasibility was fully considered.

Further, the above-mentioned prior art 6 describes the collation of the image of the defective portion with the design information, but does not describe the correction or identification of the defective position.

Generally, in a method or apparatus for irradiating charged particles such as ion particle beams other than electron beams onto a circuit board to be inspected and detecting the change occurring at that time as an image, the charged particle is to be inspected in a substrate. The same phenomenon occurs because the circuit board itself is charged positively or negatively by irradiating. It goes without saying that similar discussions can be made for these.

An object of the present invention is to solve the above problems.
Even when the charged particle beam is irradiated onto the substrate to be inspected and the charged particle beam irradiated is deflected due to charging, etc., the position coordinates of the defect can be correctly specified at the time of defect inspection or defect confirmation. To provide a pattern inspection method and an apparatus therefor.

[0016]

In order to achieve the above object, the present invention is directed to moving a stage having a reference coordinate system on which a substrate to be inspected on which a plurality of dies are arranged is mounted in a predetermined direction. Irradiating the inspected substrate with charged particles or light and sequentially detecting any of secondary electrons, backscattered electrons, transmitted electrons, and absorbed electrons generated from the inspected substrate according to the arrangement of the dies, and A / A detection step of D-converting to sequentially obtain detected digital image signals, and a reference digital image in which the detected digital image signals sequentially obtained in the detecting step are sequentially stored and delayed by the die pitch to be expected to have the same original pattern. Comparing the storage process for sequentially obtaining signals with the detected digital image signal between the dies or between the cells and the reference digital image signal, the mismatched portion is judged as a defect or a defect candidate. Then, the position coordinates of the defect or defect candidate are calculated based on the reference coordinate system, and further, the detected digital image signal and the memory which are sequentially obtained from the detection step are stored for each die in a unit of a minute region divided in the die. An image processing step of calculating a deviation amount from a reference digital image signal sequentially obtained from the steps, and a deviation amount calculated in a minute area unit for each die in the image processing step is determined from the desired die by the defect or defect candidate. Calculated in the image amount processing step and the deviation amount calculation step of calculating the deviation amount in which the defect or the defect candidate is originally located with respect to the reference coordinate system by adding up to the die in minute area units. The position coordinates of the defect or the defect candidate are corrected by the amount of deviation that is originally located in a unit of a small area in which the defect or the defect candidate is calculated, which is calculated in the deviation amount calculation step. A pattern inspection method and apparatus and having a repair step.

Further, according to the present invention, while a stage having a reference coordinate system on which a substrate to be inspected, on which a plurality of dies having a plurality of reference points whose position coordinates are known are arranged, is mounted, is moved in a predetermined direction, A / D by irradiating the substrate to be inspected with charged particles or light and sequentially detecting any of secondary electrons, backscattered electrons, transmitted electrons and absorption electrons generated from the substrate to be inspected according to the arrangement of the die. A detecting step for converting and sequentially obtaining detected digital image signals, and a die for which the detected digital image signals sequentially obtained in the detecting step are sequentially stored and delayed by a die pitch or a cell pitch to be expected to have the same pattern A storage step of obtaining a reference digital image signal according to the arrangement of FIG. And an image processing step of calculating the position coordinates of the defect or defect candidate by determining the non-matching portion as a defect or a defect candidate, and the plurality of the plurality of images obtained in the detection step. The detected digital image signal of the reference point is compared with the reference image data of a plurality of reference points whose position coordinates are known in advance to calculate the deviation amount of the detected image signal of each reference point. The deviation correction amount at the position coordinates of the defect or defect candidate calculated in the image processing step is obtained by performing interpolation (for example, linear interpolation) on the basis of the deviation amount of the defect or the defect candidate of the defect or defect candidate with the obtained deviation correction amount. A pattern inspection method and an apparatus therefor having a correction step of correcting position coordinates.

Further, the present invention is a pattern inspection method for inspecting on the basis of defect information including position information of a defect and image information of the defect on a substrate to be inspected having a circuit pattern formed thereon, Of the image information and the layout information of the substrate to be inspected, the position information of the defect is corrected by the obtained displacement amount, and the position information of the defect is the position information of the corrected defect. It is characterized by replacement or addition.

Further, according to the present invention, defect information including position information of a defect on a substrate to be inspected on which a circuit pattern is formed and image information of the defect is received, and position information of the defect in the received defect information is received. The substrate to be inspected is positioned on the basis to detect a detection image at a defect position, a deviation amount between the detected detection image and the image information of the received defect is measured, and the deviation amount is measured. A pattern defect inspection method and apparatus for correcting the received position information of a defect to confirm and inspect the defect on the inspection target substrate.

That is, according to the present invention, at the time of defect confirmation, the amount of deviation is obtained by automatically or manually comparing the image data of the pattern defect portion at the time of defect inspection with the image data detected again to confirm the defect. By correcting the position coordinates of the defect obtained at the time of defect inspection with the deviation amount, it becomes possible to move to the corrected position when moving to the next checking defect, and the next checking defect can be checked. It will be possible.

Further, according to the present invention, when the defect is confirmed, the image is moved to one of the reference points as necessary, the image of the reference point is detected, and the detected image of the reference point and the registered image of the reference point are detected. By calculating the amount of deviation between the defects and correcting the position coordinate of the defect obtained at the time of defect inspection with the calculated amount of deviation, it becomes possible to move to the corrected position when moving to the next checking defect, and It is possible to confirm the confirmation defect.

Further, the present invention receives defect information including image information of a defect position on a substrate to be inspected on which a circuit pattern is formed, detects a detection image at the defect position in the received defect information, The repetition index (the number of repetition mats) of the repetition portion (repetition mat portion) is counted on the basis of the detected image detected, and the counted repetition index is displayed or added to the defect information. A pattern defect inspection method and an apparatus therefor, which are characterized by confirming defects.

Further, according to the present invention, defect information including positions of a plurality of defects on a substrate to be inspected and image information thereof is acquired, or read from a non-volatile storage medium or read from an information transfer means and stored. A storage unit, a stage control unit that positions the inspection target substrate placed on the basis of the defect information stored in the storage unit, and an image signal on the inspection target substrate positioned by the stage control unit is detected. An image detection unit, a display unit that displays the detected image detected by the image detection unit and the image information of the defect information obtained from the storage unit in parallel or by switching or overlapping, and detected by the image detection unit. The detection amount and the displacement amount acquisition unit that acquires the displacement amount based on the image information of the defect information obtained from the storage unit, and the stage control unit based on the displacement amount obtained by the displacement amount acquisition unit. It is configured to correct the positioning of the substrate to be inspected by the detection and detect the detected image of the defect from the image detection unit and display it on the display unit. Furthermore, in the defect information displayed on the display unit, The pattern inspection apparatus is characterized by being configured so that additional information (category, classification information, cause of defect occurrence, etc.) can be added and output.

Further, according to the present invention, the defect information including the positions of a plurality of defects and the image information of the defects on the substrate to be inspected and the positions of the reference points and the dictionary image thereof are acquired or read from the nonvolatile storage medium or the information is acquired. A storage unit that reads out from the transfer unit and stores the storage unit, a stage control unit that positions the mounted inspection target substrate based on the defect information stored in the storage unit, and a positioning unit that positions the stage control unit. An image detection unit that detects an image signal on a substrate to be inspected, and a detection image detected by the image detection unit and image information of defect information obtained from the storage unit are displayed in parallel or in a switched or superposed manner. An image signal of the reference point detected by the image detection unit by moving the position of the reference point on the substrate to be inspected by controlling the stage control unit. No. and the information of the dictionary image stored in the storage unit to calculate or obtain the coordinate shift amount, and correct the defect information stored in the storage unit based on the calculated or obtained coordinate shift amount. The positioning of the substrate to be inspected by the stage control unit is corrected so that the image signal of the defect can be detected by the image detection unit and displayed on the display unit, and the defect information displayed on the display unit can be further displayed. The pattern inspection apparatus is characterized in that it is configured such that additional information about a defect (category, classification information, cause of defect occurrence, etc.) can be added and output.

[0025]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of a pattern inspection method and an apparatus therefor according to the present invention will be described with reference to specific drawings.

The pattern inspection apparatus according to the present invention is, for example, as shown in FIGS. 1, 11, and 14, a stage having a reference coordinate system on which a substrate 31 to be inspected on which a plurality of dies are arranged is mounted. 6 is a traveling controller that continuously travels in a predetermined direction, an irradiation optical system 106 that irradiates the substrate 31 to be inspected with charged particles or light, and the irradiation optical system irradiates the charged particles or light. By detecting any of light, secondary electrons, backscattered electrons, transmitted electrons, and absorbed electrons generated from the substrate to be inspected according to the array of dies in synchronization with the traveling of the traveling control unit, and performing A / D conversion. The image acquisition units 8 and 9 for obtaining the detected digital image signal and the reference data which can be expected to have the same pattern by sequentially storing the detected digital image signals sequentially obtained by the image acquisition unit and delaying them by the die pitch. A storage unit 109 that sequentially obtains digital image signals and a detected digital image signal between dies or cells and a reference digital image signal are compared with each other to determine a non-coincidence portion as a defect or defect candidate, and the position coordinates of the defect or defect candidate. Is configured based on the reference coordinate system.

(First Embodiment) A first embodiment using an electron beam of a pattern inspection apparatus according to the present invention will be described with reference to FIG. That is, in the first embodiment, an electron beam source for generating the electron beam 2 and an electron beam 2 from the electron beam source are accelerated by an electrode (not shown) to be taken out and an electrostatic or magnetic field superimposing lens (not shown) is used. ), An electron gun 102 that creates a virtual light source at a fixed location, and a condenser lens 103 that converges the electron beam 2 from the electron gun 102 that creates a virtual light source at a fixed location,
Installed near the position converged by the electron gun 102, the electron beam 2
Blanking plate 104 for controlling ON / OFF of
, A deflector 105 for deflecting the electron beam 2 in the XY directions, and an objective lens 4 for converging the electron beam 2 on the inspection target substrate 5.
And an electron optical system 106 including First
The first embodiment further includes a sample chamber 107 that holds a substrate 31 to be inspected such as a wafer in a vacuum.
In the inside 7, a substrate 6 to be inspected is mounted, and a stage 6 to which a retarding voltage 108 that enables image detection at an arbitrary position is applied is provided. The stage 6 is equipped with a laser length measuring device (not shown), and the stage 6 travels in a reference coordinate system. In the first embodiment, a detector 8 for detecting secondary electrons 7 from the substrate 31 to be inspected and a detection signal detected by the detector 8 are added to the A / D
A / D converter 9 for converting the digital image to obtain a digital image, a memory 109 for storing the digital image, a reference digital image signal stored in the memory 109 and a detected digital image for A / D conversion Then, the location having the difference is detected as the pattern defect 11, and the coordinates, projection length, area, and threshold value DD of the detected pattern defect 11 are detected.
(Limit threshold value detected as a defect when the threshold value is less than or equal to this value), the average value of the difference image between the detected digital image signal detected as a defect and the reference digital image signal, the variance of the difference image, The image processing circuit 10 includes a maximum image difference detected as a defect, a texture of a defect image, a texture of a reference image, and a feature amount 60 such as image information.

Further, even if the charge-up, which is a feature of the present invention, occurs on the substrate 31 to be inspected, the deviation amount of one die per minute area is used as a portion for calculating the original deviation amount of the pattern defect 11. Amount holding unit 60 for holding
Then, the deviation amount 61b in a minute area unit obtained by accumulating up to the preceding die from the deviation die holding unit 60 based on the reference die with almost no charge-up and the current obtained from the image processing circuit 10. And the deviation amount calculation unit 62 that calculates the deviation amounts 61c and 61d in the original minute area unit based on the deviation amount 61a in the minute area unit in the die. It should be noted that the small area unit was divided into as small pieces as possible so that the deviation of the electron beam 2 would be almost the same even if the charge-up varied in the die.

Further, as a portion for inspection processing and defect confirmation, coordinates of the pattern defect 11 detected and calculated by the image processing circuit 10, features such as projection length and area, a threshold threshold value DD, and a defect are detected. About the average value of the difference image between the detected digital image signal and the reference digital image signal, the variance of the difference image, the maximum image difference detected as a defect, the texture of the defect image, the texture of the reference image, the feature amount 60 such as image information The pattern defect storage unit 6 stores and stores the original coordinates and the like corrected by the deviation amount 61c in the original minute area unit obtained from the deviation amount calculation unit 62.
3, an overall control unit 200 that controls the entire apparatus (a control line from the overall control unit 200 is not shown), and an overall control unit 2
Display device 20 connected to 00 for displaying various operations
1, keyboard 220 and mouse 22 for instructing operation
1 and a knob 222.

Further, as a control system and an observation optical system, a Z sensor 113 for measuring the height of the wafer 31 and an offset 112 are added to the height data measured by the Z sensor 113, and this offset corrected data is used. Objective lens 4
A focus control system that keeps the focus position of the image detected by the detector 8 constant by controlling the current value of, a loader (not shown) for loading / unloading the wafer 31 in the cassette 114 into / from the sample chamber 107, and a wafer Wafer 31 based on the outer shape of 31
An orientation flat detector (not shown) for positioning the wafer 3 and the wafer 3
Optical microscope 118 for observing the pattern on
And a standard sample piece 119 provided on the stage 6. The wafer 31 is assumed to be accurately positioned with respect to the stage 6. Therefore, the wafer 31 is
It is assumed that the stage 6 travels in the reference coordinate system.

Next, the operation and action will be described.
First, the start screen shown in FIG. 2 is displayed on the operation screen 210 of the display device 201. Next, on the operation screen 210,
Wafer 3 targeted by the user in cassette 114
1 is selected by the shelf number selection component 130, the type and process of the target wafer 31 is designated by the recipe selection component 131, and the inspection start button 330 is pressed to start the inspection. The inspection is performed by loading the wafer 31 onto the stage 6, aligning and calibrating the wafer 31, then performing defect inspection processing, then performing defect confirmation processing, and after outputting the defect, remove the wafer 31 from the stage 6. Load and exit.

Here, the defect inspection process according to the present invention,
The defect confirmation process will be described. That is, the operation screen 210
In, the start button 330 is used to instruct the start of the inspection. When the inspection is started, the wafer 31 mounted by moving the stage 6 is moved to the scan start position of the region to be inspected. Since an offset unique to the wafer, which has been measured in advance, is set in the offset 112, the focus of the wafer surface is controlled by the current flowing through the objective lens 4 with the Z sensor 113 enabled. Next, the stage 6 is shown in FIG.
The scanning is performed in the Y direction along the scanning line 33 shown in FIG.
5, the electron beam 2 is scanned in a direction intersecting the Y direction (for example, the X direction) in synchronism with the stage scanning, and the voltage of the blanking plate 104 is cut off during the effective scanning to scan the electron beam 2 on the wafer 31. Irradiate. At that time, the reflected electrons or secondary electrons generated from the wafer 31 are detected by the detector 8
The digital image signal of the stripe region 34 is converted by the A / D converter 9 and stored in the memory 109. After the scanning of the stage 6 is completed, the Z sensor 113 is invalidated.

As described above, by repeating the stage scanning, the digital image signal is detected and inspected over the entire necessary area. When the entire surface of the wafer 31 is inspected, the inspection is performed in the order shown in FIG.

By the way, as shown in FIG. 4, the electron beam 2 is emitted from the reference die (first die) D ₀ 48 formed on the substrate 31 to be inspected, such as a wafer, in which charge-up has not occurred. By moving the stage 6 in the Y direction opposite to the arrow while deflecting and scanning in the direction, the detector 8 sequentially detects the detected digital image signals from the die. Note that FIG. 4 shows the main movement of the stage 6. Actually, in order to detect the image of one die by the detector 8, it is necessary to move the stage 6 a lot of times.

More specifically, as shown in FIG. 9, starting from the reference die in which charge-up has not occurred, the electron beam 2 is laterally traversed with respect to the die D _N-1 and then the die D _N. By moving the stage 6 in the direction opposite to the arrow 51 while deflecting and scanning, the detector 8 is moved to the die D _N-1.
About the unit of small area D _N-1 (1), D _N-1 (2), ……
_{..., D N-1 (n} -1), D N-1 (n), ......, detects a detection image signal from the next die D units of the sequential microscopic regions for _N D _N (1),
The detected image signal is detected from D _N (2), ..., D _N (n-1), D _N (n) ,.

As described above, when the electron beam 2 is applied to the inspected substrate 31 such as a wafer (circuit board), as shown in FIG. Secondary electrons 7 are generated. The amount of secondary electrons 7 generated is determined depending on the energy E of the irradiated electron beam 2, and a case where more secondary electrons are generated than the irradiated electron beam 2 and a small amount of secondary electrons are generated. There are cases. The ratio of the amount of generated secondary electrons 7 to the amount of the irradiated electron beam 2 is called the secondary electron emission efficiency η.
The secondary electron emission efficiency η changes depending on the energy of the electron beam 2 to be irradiated as shown in FIG. Therefore, the secondary electron emission efficiency η
When is not 1, positive or negative electrification occurs on the substrate 31 to be inspected.

By the way, when the surface of the substrate 31 to be inspected is an insulating material, or even if it is a conductor and is isolated in an insulating material or the like, the charge is accumulated. As a result, the detector 8 detects the position where the electron beam 2 to be irradiated next on the inspection target substrate 31 is deflected by this accumulated charging and the actual image and the detector 8 detects the position. Since the positions are different, the detected image signal detected by the detector 8 is displaced.

In such a state, the electron beam gun 102
The electron beam 2 from is deflected in the X direction by the deflector 105,
When the inspection target substrate 31 having a surface (for example, an insulating material or a conductor is isolated in an insulating material or the like) on which charge is accumulated is irradiated via the objective lens 4, the electron beam 2 is charged by the electron beam 2. The target substrate 3 to be inspected is irradiated with a slight deviation, and simultaneously the stage 6 is continuously moved in the Y direction.
When the detector 8 detects the secondary electrons 7 from 1 and the like, the detected image signal in which the positional deviation has occurred is detected. Then, the detected image signal in which the positional deviation has occurred is the A / D converter 9
The detected digital image signal f (x, y), which has been A / D-converted and has been displaced, is obtained and input to the memory 109 and the image processing circuit 10.

Then, in the image processing circuit 10, a reference digital image signal (for example, a signal stored in the memory 109) g (x) at a place where it can be expected to be originally the same as the detected digital image signal f (x, y). , y), and the position (| f (x, y) −g (x, y) |> Th: determination threshold value) having a difference is set as the pattern defect 11 and its coordinates (x,
y), a projection length, a feature amount such as image information, or image data is calculated and stored in the pattern defect storage unit 63. At this time, as a comparison method, "reference die comparison" shown in FIG.
There are "sequential die comparison" and "cell comparison". That is, the die reference comparison that compares the image of the reference die with the image of the die being inspected, and the sequential die that detects the images of the sequential die and compares the image of the previous die with the image of the current die. Comparison and
There is a cell comparison method in which images of repeated patterns are compared with each other by limiting to the repeated pattern portion.

First, the case of the reference die comparison method will be described with reference to FIG. In case of reference die comparison,
The charge-up can be ignored, that is, the detected digital image f (x, y) of the standard die on the wafer 31 which is first irradiated with the electron beam is used as a reference digital image signal g (x, y) in a unit of a small area ( (1), (2), ..., (n-1), (n), ...) are stored in the memory 109. As a result, in the detection of the reference die, it is assumed that charge-up is ignored and there is no deviation due to charge-up, and the deviation amount holding unit 60 is always set to 0, and the deviation amount 61a detected by the image processing circuit 10 is left unchanged. In the deviation amount calculation unit 62, the final original deviation amount 61d
Is input to the pattern defect storage unit 63 and stored as a correction value. It should be noted that it is assumed that the deviation amount holding unit 60 is always set to 0 without the deviation amount and there is no feedback signal 61c from the deviation amount calculation unit 62. As a result, in the image processing circuit 10, the reference digital image signal to be compared with the detected digital image signal is obtained from the reference die D ₀ , and the detected digital image signal and the reference digital image signal detected in the image processing circuit 10 are the same. Deviation amount 61
a is the original deviation amount for each small area unit and is directly input as a correction value (correction value) to the pattern defect storage unit 63 via the deviation amount calculation unit 62. But,
Since the reference die D ₀ that can ignore the charge-up and the die D _N currently inspected are always compared, the reference position of the reference die D _{0 and} the reference position of the die D _N currently inspected are always compared. Will need to be aligned. Therefore, the case of sequentially comparing adjacent dies described below is superior to the reference die comparison because the die pitch can be treated as constant.

Next, the case of the sequential die comparison method will be described with reference to FIG. In the case of the sequential die comparison method, the deviation amount holding unit 60 is provided before the image of the first die D ₀ is detected.
To clear. Next, the image is sequentially detected along the die of the arrow. Also, starting from the first die D _{0 in} which charge-up can be ignored and updated by the detected digital image signals of the immediately preceding die which are successively adjacent to each other, the unit of the minute area (((1 ), (2), ..., (n-1),
(n), ...) Is stored in the memory 109.

Then, the image processing circuit 10 detects the detection position D
_While detecting _N 36 (shown in FIG. 3), the A / D converter 9
The detected digital image signal of the detection position D _N 36 detected from the detection position D _N-1 35 stored in the memory 109.
The reference digital image signal (shown in FIG. 3) is compared, and the difference is extracted as the pattern defect 11 and
The positional deviation amount 61a for each micro area between the detection position D _N 36 and the detection position D _N-1 35 is sent to the deviation amount calculation unit 62.
In the image processing circuit 10, the pattern defect 11 is extracted, and the detection position D _N 36 and the detection position D _N-1 3 are detected.
The method of obtaining the positional deviation amount 61a for each minute area unit between No. 5 and No. 5 is described in Japanese Patent Application Laid-Open No. 10-318950, and will be omitted. At the same time, the detected digital image signal at the detection position D _N 36 detected by the A / D converter 9 is stored in the memory 109 to obtain the next reference digital image signal.

The deviation amount calculation unit 62 obtains the reference die D _{0 having} substantially no charge-up obtained from the address of each minute area of the deviation amount table which constitutes the deviation amount holding unit 60.
To the die D _{N-1 in} which the detection position D _N-1 35 exists, the deviation amount 61b accumulated for each minute area unit and each minute amount of the die D _N with respect to the die D _N-1 obtained from the image processing circuit 10. The original deviation amounts 61c and 61d for each minute area unit in the die B are calculated by adding the deviation amount 61a for each area unit, and the calculated deviation amounts 61c for each minute area unit in the die B are calculated. The shift amount table of the shift amount holding unit 60 is stored at the address of each small area unit. as a result,
At each address of the deviation amount table of the deviation amount holding unit 60,
Amount of deviation for each original micro area accumulated up to die D _{N based} on reference die D ₀ that can ignore charge-up 6
1c will be retained. At the same time, the shift amount calculation unit 6
From 2, the original deviation amount 61d for each micro area is input to the pattern defect storage unit 63, and the above-mentioned coordinates with respect to the coordinates of the defect pattern 11 detected for each micro area in the die D _N from the image processing circuit 10 are described above. The image processing circuit 10 is corrected by the input original deviation amount 61d for each minute area.
The corrected coordinates are stored in the pattern defect storage unit 63 together with the defect digital image signal and the defect feature amount obtained from the above.

The coordinates of the defect pattern 11 detected by the image processing circuit 10 are detected based on the movement amount of the stage 6 which is a reference coordinate system measured by a laser length measuring device (not shown). Further, it is assumed that the pitch of the arrayed dies is constant. However, when the die pitch varies, the reference mark or reference point formed for each die is detected by the detector 8.
Die pitch deviation correction by detecting by
That is, the pitch deviation between the detected digital image signal obtained from the A / D converter 9 and the reference digital image signal read from the memory 109 is corrected.

That is, the coordinates of the defect pattern 11 detected by the image processing circuit 10 are the reference marks formed for each die with respect to the movement amount of the stage 6 measured by a laser length measuring device (not shown). Alternatively, it is obtained by being corrected based on the die pitch shift obtained by detecting the reference point.

That is, in the case of sequential die comparison, the deviation amount holding unit 60 is cleared before the inspection is started, and the inspection starts from the reference die D _{0 in} which charge-up can be ignored and continues to the die D _N to be inspected. In doing so, the deviation amount for each small area unit accumulated up to the preceding die D _N-1 is held in the deviation amount holding unit 60. That is, as shown in FIG. 8, the comparison deviation amount calculation unit 62 calculates the deviation amount holding unit 60.
Accumulated deviation amount ΣΔD for each minute area unit held in
_N-1 (n) and the reference digital image signal g (x, y) of the micro area unit (D _N-1 (n)) of the previous die D _N-1 obtained from the image processing circuit 10 are detected. A deviation amount (ΔD _N (n) = D _N (n) −D _N−1 ) between the detected digital image signal f (x, y) of the micro area unit (D _N (n)) of the die D _N to be processed. (n)) and the accumulated deviation amount (ΣΔD _N (n)) 61 in the die D _N
c and 61d are calculated for a range of one die ((1) to (n)) and are held in the shift table of the shift amount holding unit 60 and are input to the pattern defect storage unit 63 to correct the coordinates of the pattern defect. Will be done. In addition, the previous die D
_N-1 micro area unit (D _N-1 (n)) reference digital image signal g (x, y) and detected micro area unit of die D _N (D
The deviation amount (ΔD _N (n)) between _N (n)) and the detected digital image signal f (x, y) is, for example, 1024 in a minute area unit.
When the number of pixels is about the same, it is an average value of the shift amounts over the pixels.

As a result, even if charge up occurs on the surface of the wafer 31 by irradiating the electron beam 2 on the dies that are in the process of being sequentially inspected, the charge up is negligible with reference to the reference die D _0. By accumulating the deviation amount for each area unit up to the die being inspected, the original deviation amount for each micro area unit in the inspection die is obtained, and the deviation amount for each original micro area unit in the inspection die thus obtained is used to calculate the image. Absolute coordinates set on the wafer 31 measured by the laser length-measuring device (not shown) by correcting the coordinates of the pattern defect detected in the inspection die from the processing circuit. And the image of the pattern defect, together with the pattern defect storage unit 6
3 will be stored.

Next, the case of the cell comparison system will be described. In the case of cell comparison, cell comparison may be used to detect a pattern defect, and the reference die comparison method or the sequential die comparison method described above may be used for deviation correction (deviation correction).
That is, in the case of cell comparison, a sequential die comparison method,
Alternatively, the position is corrected by performing at least alignment by the reference die comparison method. Further, by performing die comparison and cell comparison at the same time as an inspection method, it is possible to correct the defect position of the cell comparison by the deviation amount of the die comparison.

As described above, the image processing circuit 1 is operated in a state where charge-up occurs in each die.
Features such as coordinates of the pattern defect 11 calculated from 0, projection length, area, limit threshold value DD, difference image average value, difference image variance, maximum image difference, defect image texture, reference image texture, image information The quantity information is stored in the pattern defect storage unit 63.
However, the defect position when the charge-up occurs is obtained from the deviation amount calculation unit 62 for each original minute area in absolute coordinates measured by the laser length-measuring device in which the charge-up does not occur. It will be corrected by the deviation amount 61d. Then, various pattern defect information 60 stored in the pattern defect storage unit 63 is input to the overall control unit 200.

Next, an embodiment of the defect confirmation processing will be described. That is, after the inspection of the required area is completed, the overall control unit 200 displays the GUI initial screen 211 shown in FIG.
The GUI defect confirmation screen 212 shown in is displayed on the display device 201. On the initial screen 211 shown in FIG. 9, a recipe creation item selection button 142 is displayed. The recipe creation item is selected depending on which button 142 is pressed, and the inspection data of the target wafer type or process is selected. Will be. The recipe save button 134 is a button for saving the selected recipe creation item. The recipe creation end button 133 is a button indicating that the recipe creation is completed. The initial threshold value setting component 145 is for setting an initial threshold value for determining a pattern defect for the difference image in the image processing circuit 10 at the start of inspection. The map display unit 55 also displays the distribution of the pattern defects 11 detected for a certain wafer at the original position coordinates. In FIG. 9, for example, the distribution of pattern defects is displayed only for the dies indicated by diagonal lines.

GUI defect confirmation screen 212 shown in FIG.
Is a defect display editing part 150 capable of displaying the feature amount of the defect and editing the classification, a current position display 59 showing the current position, and a pattern display 11 showing the pattern defect 11 displaying the classification number as a symbol together with the layout information of the wafer 31. 55, an image display unit 56 displaying the image at the current position, a display switching button 151 for switching the map display format, and an inspection end button 144 for instructing the end of inspection.

When the mouse operation instruction button 140 is set to the selection mode and the pattern defect 11 is clicked, the overall control unit 200 causes the pattern defect storage unit 63 shown in FIG.
Defect inspection image or pattern defect storage unit 63 stored in
The confirmation defect image detected by moving again to the confirmation defect position stored in is displayed on the image display unit 56, and the feature amount of the pattern defect (corrected original coordinate value, projection length (X
Axis, Y axis, rotational symmetry axis, etc.), area, etc.) are displayed on the defect display edit component 150.

The overall control unit 200 controls the image display unit 5
Defect image for confirmation and defect display editing part 15 displayed in 6
The pattern defect 11 is classified based on the defect feature amount displayed at 0, and the defect display editing component 150 assigns a classification number (additional information about the defect) to the feature amount of the pattern defect 11.

As described above, when the defect is reviewed (confirmed), the position of the defect is irradiated with the electron beam 2 again, but if the vicinity of the defect is charged, the electron beam 2 is deflected and detected. A deviation will occur in the confirmation defect image.

Therefore, the overall control unit 200 moves again to the confirmation defect position stored in the pattern defect storage unit 63 for confirmation, detects the confirmation defect image by the detector 8, and A / D converts it. When the defect image (confirmation defect image) is detected again from the image processing circuit 10 after A / D conversion by the container 9, the detected confirmation defect image and the high-speed image before being stored in the pattern defect storage unit 63. The defect inspection image at the time of inspection is displayed on the image display unit 56 at the same time, and the operator is instructed of the deviation amount, or the position is automatically adjusted to calculate the deviation amount, or the deviation amount is automatically calculated. Either presenting the calculated deviation amount to the operator and correcting it as necessary, or specifying the deviation amount of the defect position using the detection image (confirmation image) without using the defect inspection image, etc. Calculate the amount of coordinate shift by Therefore, this coordinate shift amount is determined. The determined coordinate shift amount is the shift amount caused by charge-up at the time of defect confirmation (at the time of defect review). Reference numeral 141 is a button for switching the defect image displayed on the image display unit 56.

Next time, for the purpose of confirmation, when the defect image for confirmation is detected by moving to the position of the next confirmation defect (review defect), the next defect for confirmation stored in the pattern defect storage unit 63 is detected. The inspection position data is corrected by the determined deviation amount, and the wafer 31 is moved to the corrected position to detect a confirmation defect image at the original position and display the confirmation defect image. It can be displayed on the part 56 and can be confirmed. That is, at the time of defect confirmation, the image data of the pattern defect portion 11 which is stored in the pattern defect storage portion 63 and which is inspected at high speed, and the defect pattern (review defect) are again irradiated with an electron beam to confirm the image data. The overall control unit 200 compares the image data detected by 8 automatically or manually to obtain the deviation amount, and the deviation amount thus obtained is used as the position coordinate of the defect obtained during the defect inspection stored in the pattern defect storage unit 63. By correcting the defect for the next confirmation, the defect can be moved to the corrected position at the time of moving to the next defect for confirmation, and the defect for the next confirmation can be confirmed.

Next, the inspection end button 144 is used to end the defect confirmation, and the result is stored in a nonvolatile storage medium (not shown) or in FIG.
After outputting to the information transfer means 500 or the like shown in 3, the initial screen is returned.

According to the first embodiment, at the time of inspection, in a minute area unit (n) up to the die D _N-1 adjacent to the inspection die D _N with reference to the reference die D ₀ or the like in which charge-up can be ignored. Since the accumulated shift amount ΣΔD _N-1 (n) is always held in the shift amount holding unit 60, the shift amount calculated by the image processing circuit 10 (ΔD _N (n) = D _N (n) -D) _N
Since the defect position can be corrected by using _-1 (n), even if the charge is charged by irradiating the inspection die and the die in the vicinity thereof with the electron beam 2, the position of the pattern defect 11 is more accurate. There is a feature that can be coordinate. However, there is a possibility that an integration error may occur due to the fact that the deviation amount is integrated up to the inspection die based on the die in which the charge-up can be ignored. The position coordinates of the generated defect can be obtained. Further, since the pattern defect storage unit 63 holds the image information of the pattern defect 11, it is possible to correct a position error due to different charge-up at the time of defect inspection and at the time of defect confirmation.

Further, since the pattern defect storage unit 63 holds the image information of the pattern defect 11, even if the defect is confirmed a plurality of times, the position error due to charge-up in each case can be corrected. There is a feature that can be.

When the defect is confirmed, the image is detected by moving to the position of the pattern defect 11, and the defect position can be corrected based on the detected image. Therefore, the defect inspection and the defect confirmation can be performed without using the image at the defect inspection. There is a feature that position errors due to different charge-ups can be corrected. Further, when the defect is confirmed, the pattern is moved to the position of the pattern defect 11, the image is detected, and the defect position can be corrected based on the detected image and the image at the time of inspection. There is a feature that can correct the error of.

Next, a first modification of the first embodiment will be described. When confirming a defect, the overall control unit 200 detects an image with a low magnification (widens the deflection width of the electron beam 2 or widens the traveling width of the stage 6) around the defective portion, or detects an image with an optical microscope. The information of the generated defect is added to the above-mentioned defect information, and the nonvolatile storage medium or the information transmitting means is provided.
Defect information is passed to another device (for example, analysis device or classification device) via (network). With another device, the position can be corrected automatically by referring to the image information added to the defect information when moving to the defect portion and performing matching, or while referring to the image. For example, as a detection image, 0.01 μm to 0.5
Assuming that a JPEG-compressed image with 1024 pixels for μm pixels is used, a capacity of about 100 kbytes will result in 10 μm-50
Images of 0 μm square area can be saved. It is considered that the area of this extent has sufficient resolution in both the peripheral circuit section and the memory mat section and that there is a reference pattern in the image. According to this modification, even if the defect position is shifted due to charge-up or the like at the time of defect inspection or defect confirmation, the defect position can be specified based on the image of the defect portion. Further, in the case of the memory mat portion, the number of memory cells from the end of the memory mat can be added as reference information with reference to the repetition frequency of the memory cells. According to this modification, since the memory cell can be specified, even if the defect cannot be specified as an image, the location can be specified based on the number of memory cells. Further, the image at the time of inspection may be stored instead of the image at the time of confirming a defect. According to this modification, there is a feature that the defect position can be specified by another device without confirming the defect.

A second modification of the first embodiment will be described. Image information of a defective portion at the time of inspection is stored and collated with design information to specify a defect position on the design information. This allows you to save the defect location in a format that allows you to accurately recognize the defect location.
Or it can be displayed. According to this modification, since the design information is referred to, the absolute position of the defect can be specified.

Next, a second embodiment using the electron beam of the pattern inspection apparatus according to the present invention will be described with reference to FIGS. 11 and 12. In the second embodiment,
The difference from the first embodiment is that the image processing circuit 10 ',
The reference data storage unit 40, the shift amount calculation unit 41, and the shift interpolation unit 42.

That is, the start of inspection is instructed by the inspection start button 330 shown in FIG. When the inspection is started, the stage 6 moves to the scanning start position of the area of the mounted wafer 31 to be inspected. In the focused state where the Z sensor 113 is enabled, the stage 6 is scanned in the Y direction along the scanning line 33 shown in FIG. Direction) and the electron beam 2 is scanned and irradiated on the wafer 31 during effective scanning. Reflected electrons or secondary electrons generated from the wafer 31 are detected by the detector 8 and a digital image of the stripe region 34 is obtained by the A / D converter 9 and stored in the memory 109. After scanning the stage 6, the Z sensor 113 is invalidated. By repeating the stage scanning, the entire surface of the necessary area is inspected. Wafer 3
When the entire surface of No. 1 is inspected, the inspection is performed in the order shown in FIG.

[0065] First, prior to the image detection of the first die D ₀ 48 shown in FIG. 4, a test area is repeated in the reference data holding unit 40 die _{D ···, D N-1,} D N, D ... Reference points (reference marks) r1 to r8 and their dictionary images (registered dictionary)
Set and register Ir1 to Ir8. The coordinates at which the reference points (reference marks) r1 to r8 provided in each die are formed are determined with reference to each die. That is, reference points (reference marks) r1 to r8 provided in each die
Positional coordinates Ir1 (x, y) of the dictionary images Ir1 to Ir8 about
Ir8 (x, y) is predetermined with respect to a reference coordinate system for the wafer 31 obtained from a laser length measuring device (not shown) that measures the movement of the stage 6. That is, in the reference die D ₀ or the like in which charge-up can be ignored, the position coordinates of the dictionary image at the reference points (reference marks) r1 to r8 can be detected at the original position by the detector 8.

Next, at the inspection stage, the stage 6 is moved, the electron beam 2 is scanned and irradiated by the deflector 105, and the detector 8 sequentially detects the image signal along the die of the arrow. . While detecting the detection position D _N 36 (shown in FIG. 3), the image processing circuit 10 causes the A / D converter 9
The detection digital image signal of the detection position D _N 36 obtained from the detection position D _N-1 35 (FIG. 3) stored in the memory 109.
The reference defect image is compared with the reference digital image signal in (1) to extract a location having a difference as a pattern defect 11, and the feature amount such as coordinates, projection length, and area of the extracted pattern defect 11 is calculated to calculate a pattern defect storage unit. The image data 43a in the vicinity of the coordinates of the reference points (reference marks) r1 to r8 existing on the die D _N is sent to the shift amount calculation unit 41. At the same time, the image signal at the detection position D _N 36 obtained from the A / D converter 9 is stored in the memory 109.

In the shift amount calculating section 41, for example, as shown in FIG. 12, the detector 8 detects reference points (reference marks) r1 to r8 in the die D _N and the A / D converter 9 performs A / D conversion. Based on the converted image data (fr1 (x, y) to fr8 (x, y)) 43a and the coordinates (x, y) of the die input from the image processing circuit 10, the reference data holding unit 40 outputs the data. Dictionary image (I
Based on r1 (x, y) to Ir8 (x, y)) 43b, the shift amount (r1 (δ
x, δy) = fr1 (x, y) −Ir1 (x, y), ..., r8 (δx, δy)
= Fr8 (x, y) -Ir8 (x, y)) 43c is calculated. The deviation interpolation unit 42 calculates the deviation amount (r1 based on the charge-up at the reference point (reference mark) position calculated by the deviation amount calculation unit 41.
Based on (δx, δy) to r8 (δx, δy)) 43c, the position d (x, y) 44 of the pattern defect 11 obtained from the image processing circuit 10 is obtained.
The displacement amount at is obtained by linear interpolation, for example, and the obtained displacement amount 45 is input to the pattern defect storage unit 63 to correct the position coordinates of the pattern defect 11.

The coordinates of the pattern defect 11 extracted by the image processing circuit 10, the projection length, the area, the critical threshold value DD (threshold value detected as a defect when the threshold value is less than this value),
Feature amount information such as difference image average value, difference image variance, maximum image difference, defect image texture, reference image texture, and image information is stored in the pattern defect storage unit 63, and the original deviation amount 45 at the defect position at that time is stored. Is added to correct the defect position. The overall control unit 200 displays the defect confirmation screen shown in FIG. 10 after the inspection of the necessary area is completed.

Confirmation and classification of defects are the same as in the first embodiment.

That is, the operator moves to a reference point (reference mark) in the die, detects the reference point detection digital image signal fr (x, y), and detects the dictionary image Ir (x,
y) and a shift amount r (δx, δy) is calculated, and the shift amount of the position coordinates of the checking defect detected from the image processing circuit 10 with respect to the checking defect is obtained by, for example, linearly interpolating between the reference points. Determine. However, when the image is detected by moving to the defect position next time, it moves to the position corrected by the determined shift amount.

The defect confirmation is ended by the inspection end button, the result is output, and the screen returns to the initial screen.

Further, according to the second embodiment, the position coordinates of the pattern defect are obtained by, for example, performing linear interpolation based on the absolute position of the deviation amount at the reference point (reference mark) r formed in the die. The feature is that the pattern defect can be specified by the absolute position because it can be corrected. However, if the number of reference points (reference marks) formed in the die is small,
For example, the accuracy may decrease due to the linear interpolation. However, it is possible to obtain the position coordinates of the defect originally generated on the wafer 31 by correcting the deviation of the defect due to charge-up.

According to the second embodiment, the first
Similar to the embodiment described above, since the pattern defect storage unit 63 holds the image information of the pattern defect 11, it is possible to correct the position error due to different charge-up at the time of defect inspection and at the time of defect confirmation. There is a feature that can be done.

Next, a third embodiment of the overall system according to the present invention will be described. FIG. 13 shows the configuration of the entire system constructed on the network. The entire system is provided with a server 501 on a network 500, and a plurality of inspection devices (an inspection device A 502 and an inspection device B in the figure).
503), the review device 504, and the result confirmation device 5
It consists of 05. Although each of the review device 504 and the result confirmation device 505 has a single configuration, it may have a plurality of configurations. Inspection device A502 or inspection device B5
From 03, the information of the pattern defect 11 including the image information of the defect portion is transferred to the server 501. Here, the positioning of the review device 504 and the result confirmation device 505 will be described.
The review device 504 is a device that receives information on the pattern defect 11 from the inspection device A 502 and the wafer 31 to be inspected, and observes defects on the wafer based on the information on the pattern defect 11, and an optical detection system. Formula, electron beam type,
There are devices for observing FIB or the like, or various analyzers.

On the other hand, the result confirmation device 505 receives only the information of the pattern defect 11 (including the defect feature amount),
It is a device that analyzes the results. The analysis of this result may be performed by using only the information of the pattern defect 11 of one wafer 31 or by using the information of a plurality of wafers, or by further analyzing the information together with the design information.

Further, the result confirmation device 50 is installed on the server 501.
The defect information obtained from No. 5 is collated with the design information to specify the absolute position of the defect, and the coordinate information of the specified defect is added to the information of the pattern defect 11 and transferred to the review device 504. Further, the server 501 transfers the defect image information including the defect image information to the review device 504 or the result confirmation device 505,
The operator identifies the defect position by referring to the defect image information in the review device 504 or the result confirmation device 505. In addition, the pattern defect 1 transferred from the server 501
The image information in the information 1 is aligned with the image information detected by the review device 504, and the original defect position is marked and displayed on the review image. Further, in the review device 504, the image information and the design information in the information of the pattern defect 11 are displayed on the screen at the same time, or switched, or overlapped. According to the third embodiment,
There is a feature that the defect information can be visually confirmed on the review device 504, or can be confirmed including the design information. In the result confirmation device 505, image information is displayed on the screen as the information of the pattern defect 11 and the layout information in the design information is also displayed. The absolute position of the defect is specified by a manual or automatic method by position alignment on the image. According to the third embodiment, there is a feature that the defect position can be specified on the screen or the image.

As described above, in the review device 504, the cause of the defect is identified by visually observing the information on the defect, classifying the defect into categories, and specifying the place where the defect occurs. It is possible to estimate.

Next, a fourth embodiment using light of the pattern inspection apparatus according to the present invention will be described with reference to FIG.

FIG. 14 shows the configuration of the optical inspection device according to the fourth embodiment. In the fourth embodiment,
The difference from the first embodiment is that the light source 21 and the light source 2
Wafer 3 which is the substrate to be inspected through the half mirror (number not set) of ultraviolet light such as excimer laser light from 1
1, an objective lens 22 that converges the light on the first surface, and a one-dimensional image sensor 23 that detects reflected light from the wafer 31 and T
Two-dimensional image sensor 45 composed of DI sensor 45
0, the image sensor 23 and the two-dimensional image sensor 4
And a switch 451 for switching 50 signals. In the case of the fourth embodiment, what is irradiated onto the wafer 31 is the ultraviolet light, so that the phenomenon that electric charges are accumulated does not occur. However, the illumination method is the same as illuminating the inspection object 31 with the electron beam (electron beam) 2.
The stripe region 34 is illuminated by moving the stage 6 on which the substrate 31 to be inspected is moved in the direction indicated by the scanning line 33 in FIG. Alternatively, the stripe region 34 is illuminated by irradiating a slit-shaped ultraviolet light beam having the width of the stripe region 34.

Next, the operation will be described. SE here
Only the part that is different from the M formula will be described. To check for defects, move the stage 6 to that location and
The focus position is adjusted by making the sensor 113 effective and keeping the position of the objective lens 22 constant, that is, the detection value of the Z sensor 113 + the offset 112. The switch 451 is switched to the two-dimensional image sensor 451, and the signal detected by the image sensor 450 is converted into a digital image by the A / D converter 9. The inspection operation moves the stage 6 and mounts the wafer 3
1 to the scanning start position of the area to be inspected. The offset unique to the wafer, which has been measured in advance, is set to the offset 112, the Z sensor 113 is enabled, and the switch 451 is switched to the image sensor 23. The stage 6 is scanned in the Y direction along the scanning line 33 shown in FIG.
The image sensor 23 detects the reflected light, the A / D converter 9 obtains a digital image of the stripe region 34, and the memory 10
Store in 9. After the scanning of the stage 6 is finished, the Z sensor 113
Invalidate. By repeating the stage scanning, the entire surface of the necessary area is inspected. When inspecting the entire surface of the wafer 31, the inspection is performed in the order shown in FIG. According to the fourth embodiment, although there is no influence due to charge-up, the defect position at the time of observation or the absolute position can be specified optically, so that a defect type different from the SEM method can be detected. There is.

[0081]

According to the present invention, when a substrate to be inspected is irradiated with a charged particle beam or the like, even if the substrate to be inspected is charged, the charged particle beam or the like is deflected. Since it is possible to correct the deviation of the defect position coordinates or the deviation of the coordinates at the time of defect confirmation, it is possible to specify minute defects with high accuracy and add various information such as categories, classifications and causes of defects. It will be possible.

[Brief description of drawings]

FIG. 1 is a first electron beam type pattern inspection apparatus according to the present invention.
It is a figure which shows the structure of embodiment of this.

FIG. 2 is a diagram showing an operation screen of the display device shown in FIG.

FIG. 3 is a diagram showing a stripe region by a scanning line with respect to a wafer layout.

FIG. 4 is a diagram showing an inspection order at the time of full surface inspection.

FIG. 5 is a diagram for explaining a charging phenomenon.

FIG. 6 is a diagram for explaining secondary electron emission efficiency.

FIG. 7 is a diagram showing an inspection method.

FIG. 8 is an explanatory diagram in which, in the first embodiment shown in FIG. 1, deviation amounts are sequentially accumulated for each die, starting from a die in which charge-up can be ignored, in units of minute regions divided in the die. is there.

FIG. 9 is a diagram showing a GUI inspection initial screen according to the present invention.

FIG. 10 is a diagram showing a defect confirmation screen of a GUI according to the present invention.

FIG. 11 is a diagram showing a configuration of a second embodiment of an electron beam type pattern inspection apparatus according to the present invention.

FIG. 12 is a diagram illustrating a first embodiment shown in FIG. 11, in which a shift amount is calculated based on a correlation between a digital image signal detected from a reference point (reference mark) formed in a die and a registration dictionary; It is a figure for demonstrating the system which calculates | requires the shift amount in a defect detection position by linearly interpolating the calculated shift amount.

FIG. 13 is a diagram showing a configuration of an inspection system according to the present invention.

FIG. 14 is a diagram showing a configuration of an optical pattern inspection device according to the present invention.

[Explanation of symbols]

2 ... Electron beam, 4 ... Objective lens, 5 ... Substrate to be inspected, 6
… Stage, 7… Secondary electron, etc. 8… Detector, 9… A / D
Converter, 10 ... Image processing circuit, 11 ... Pattern defect candidate, 21 ... Light source, 22 ... Objective lens, 23 ... Image sensor, 31 ... Wafer (inspected substrate), 32 ... Die,
33 ... Scan line, 34 ... Stripe region, 35 ... Detection position D _N-1 , 36 ... Detection position D _N , 40 ... Reference data holding unit,
41 ... Deviation amount calculation unit, 42 ... Deviation interpolation unit, 48 ... First die (reference die) D ₀ , 55 ... Map display unit, 56 ... Image display unit, 59 ... Current position, 60 ... Deviation amount holding unit, 62 …
Deviation amount calculation unit, 63 ... Pattern defect storage unit, 102 ... Electron gun, 103 ... Condenser lens, 104 ... Blanking plate, 105 ... Deflector, 106 ... Electron optical system, 1
07 ... sample chamber, 108 ... retarding voltage, 109 ...
Memory, 112 ... Offset, 113 ... Z sensor, 11
4 ... Cassette, 118 ... Optical microscope, 119 ... Standard sample piece, 200 ... Overall control unit, 201 ... Display device, 210
... Operation screen, 211 ... GUI initial screen, 212 ... GUI
Defect confirmation screen, 220 ... Keyboard, 221 ... Mouse,
222 ... knob, 130 ... shelf number selection part, 131 ... recipe selection part, 132 ... recipe creation start part, 133 ... recipe creation end button, 134 ... recipe save button, 14
0 ... Mouse operation instruction button, 141 ... Image switching button,
142 ... Recipe creation item selection button, 143 ... Inspection start button, 144 ... Inspection end button, 145 ... Initial threshold setting component, 150 ... Defect display editing component, 151 ... Display switching button, 160 ... New area creation button, 161 ... Finish button, 200 ... Image processing area, 330 ... Inspection start button, 500 ... Network, 501 ... Server, 502 ...
Inspection device A, 503 ... Inspection device B, 504 ... Review device, 505 ... Result confirmation device.

Continued front page    (72) Inventor Maki Tanaka             292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa             Inside the Hitachi, Ltd. production technology laboratory (72) Inventor Asahiro Kunibe             292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa             Inside the Hitachi, Ltd. production technology laboratory (72) Inventor Chie Shishido             292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa             Inside the Hitachi, Ltd. production technology laboratory (72) Inventor Hidetoshi Nishiyama             292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa             Inside the Hitachi, Ltd. production technology laboratory (72) Inventor Hiroto Okuda             292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa             Inside the Hitachi, Ltd. production technology laboratory (72) Inventor Hiroshi Miyai             882 Ichimo, Hitachinaka City, Ibaraki Stock Association             Company Hitachi Ltd. measuring instrument group (72) Inventor Yasuhiro Gunji             882 Ichimo, Hitachinaka City, Ibaraki Stock Association             Company Hitachi Ltd. measuring instrument group (72) Inventor Atsushi Shimoda             292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa             Inside the Hitachi, Ltd. production technology laboratory (72) Inventor Yoshimasa Oshima             292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa             Inside the Hitachi, Ltd. production technology laboratory F-term (reference) 4M106 AA01 BA02 CA39 DB04 DB05                       DB12 DB18 DB30 DJ02 DJ04                       DJ05 DJ18 DJ20 DJ21 DJ23

Claims (21)

[Claims] 1. A / D by irradiating a substrate to be inspected on which a circuit pattern is formed with charged particles or light, and detecting any of the generated light, secondary electrons, reflected electrons, transmitted electrons, and absorbed electrons.
An image detecting step of converting to obtain a detected digital image signal; and a detected digital image signal obtained in the image detecting step,
A defect detection step of detecting a defect or a defect candidate based on the mismatch and comparing the position coordinates thereof with a reference digital image signal that can be expected to have originally the same circuit pattern, and referring to the detected digital image signal A correction step of calculating a deviation amount between the two signals when comparing with the digital image signal, and correcting the position coordinates of the defect or defect candidate calculated in the defect detection step based on the calculated deviation amount. A method for inspecting a pattern, which comprises: 2. The substrate to be inspected is irradiated with charged particles or light while traveling in a predetermined direction on a stage having a reference coordinate system on which the substrate to be inspected on which a plurality of dies are arranged is placed. A detection step of sequentially detecting any of secondary electrons, backscattered electrons, transmitted electrons, and absorbed electrons generated from the substrate to be inspected according to the array of dies, and A / D converting the detected digital image signals. A storage process of sequentially storing the detected digital image signals sequentially obtained in the detection process and delaying them by the die pitch to sequentially obtain a reference digital image signal that can be expected to have the same pattern, and a detection digital signal between dies or cells. The image signal and the reference digital image signal are compared to determine the non-coincidence portion as a defect or defect candidate, and the position coordinates of the defect or defect candidate are set to the reference The amount of deviation between the detected digital image signal sequentially obtained from the detection step and the reference digital image signal sequentially obtained from the storage step is calculated for each die in units of minute areas obtained by dividing the inside of the die. The image processing step to be calculated, and the deviation amount calculated for each die in the minute area unit in the image processing step is integrated in the minute area unit from the desired die to the die for which the defect or the defect candidate is determined. A deviation amount calculating step of calculating a deviation amount in which a defect candidate is originally located with respect to the reference coordinate system in a unit of a small area, and a position coordinate of the defect or the defect candidate calculated in the image processing step, the deviation amount calculating step. A pattern inspecting method, which comprises a correction step of correcting with a deviation amount that is originally located in a unit of a minute area in which the defect or defect candidate calculated in . 3. A charged particle or light is irradiated onto the substrate to be inspected while a stage having a reference coordinate system on which a substrate to be inspected on which a reference die is formed is mounted is run in a predetermined direction. A second electron, a backscattered electron, a transmitted electron, or an absorbed electron generated from the reference die of the substrate to be inspected is detected and A / D converted to obtain and store a reference digital image signal of the reference die. 1) irradiating the inspected substrate with charged particles or light while traveling in a predetermined direction on a stage having a reference coordinate system on which the inspected substrate on which a plurality of dies are arranged is placed. Then, any of secondary electrons, backscattered electrons, transmitted electrons, and absorbed electrons generated from the substrate to be inspected according to the array of dies is sequentially detected and A / D converted to obtain a detected digital image signal according to the array of dies. The second inspection to get The output step and the detected digital image signal obtained in the second detection step and the reference digital image signal stored in the first detection step are compared to determine the mismatched portion as a defect or a defect candidate. An image processing step of calculating the position coordinates of the defect or the defect candidate based on the reference coordinate system, and further calculating a deviation amount of the detected digital image signal with respect to the reference digital image signal; And a correction step of correcting the calculated position coordinates of the defect or the defect candidate with the shift amount calculated in the image processing step. 4. The pattern according to claim 2, wherein when the position coordinates of the defect or defect candidate are calculated in the image processing step, a feature amount including image information of the defect or defect candidate is also calculated. Inspection method. 5. A / D by irradiating a substrate to be inspected on which a circuit pattern is formed with charged particles or light and detecting any of the generated light, secondary electrons, reflected electrons, transmitted electrons, and absorbed electrons.
An image detecting step of converting to obtain a detected digital image signal; and a detected digital image signal obtained in the image detecting step,
A defect or defect candidate detecting step of detecting a defect or a defect candidate based on the inconsistency by comparing with a reference digital image signal which can be expected to be originally the same circuit pattern and calculating position coordinates thereof, and the image detecting step. The detected digital image signals of a plurality of reference points formed on the substrate to be inspected are compared with reference image data of a plurality of reference points whose position coordinates are known in advance, and the reference image data of each reference point is compared. The shift amount of the detected image signal is calculated, and the shift correction amount in the position coordinates of the defect or the defect candidate calculated in the defect detection step is calculated by interpolating based on each of the calculated shift amounts. A pattern inspection method comprising: a correction step of correcting the position coordinates of a defect or a defect candidate with a deviation correction amount. 6. A stage having a reference coordinate system on which a substrate to be inspected, on which a plurality of dies having a plurality of reference points whose position coordinates are known are arrayed, is placed, while traveling in a predetermined direction,
Irradiating the substrate to be inspected with charged particles or light,
A detection step of sequentially detecting any of secondary electrons, backscattered electrons, transmitted electrons, and absorbed electrons generated from the substrate to be inspected according to the arrangement of the dies, and A / D converting the detected digital image signals sequentially; A storage step of sequentially storing the detected digital image signals sequentially obtained in the step and delaying them by a die pitch or a cell pitch to obtain a reference digital image signal in accordance with an array of dies which can be expected to have originally the same pattern; The detected digital image signal obtained in the detection step and the reference digital image signal obtained in the storage step are compared to determine the non-coincidence portion as a defect or a defect candidate, and the position coordinates of the defect or the defect candidate are set to the reference. An image processing step of calculating based on a coordinate system, and pre-registering with the detected digital image signals of the plurality of reference points obtained in the detecting step. The image is calculated by comparing the reference image data of a plurality of reference points whose position coordinates are known to calculate the amount of deviation of the detected image signal of each reference point, and interpolating based on each of the calculated amounts of deviation. A pattern inspection comprising: a deviation correction amount at the position coordinates of the defect or defect candidate calculated in the processing step; and a correction step of correcting the position coordinates of the defect or defect candidate by the calculated deviation correction amount. Method. 7. A / D by irradiating a substrate to be inspected on which a circuit pattern is formed with charged particles or light and detecting any of the generated light, secondary electrons, reflected electrons, transmitted electrons, and absorbed electrons.
The detected digital image signal is converted to obtain a detected digital image signal, and the obtained detected digital image signal is compared with a reference digital image signal that can be expected to have originally the same circuit pattern to detect a defect or a defect candidate based on a mismatch. Defect inspection step of calculating the position coordinates of the inspection target substrate, and the inspection target substrate is moved and positioned based on the position coordinates of the confirmation defect calculated in the defect inspection process, and the positioned inspection target substrate When irradiating the upper part with charged particles or light and detecting any of the generated light, secondary electrons, backscattered electrons, transmitted electrons, and absorbed electrons and performing A / D conversion to obtain a digital image signal for confirmation, The confirmation digital image signal is automatically compared with the detection digital image signal detected in the defect inspection step, or the confirmation digital image signal and the detection digital image signal are compared. The displacement amount is confirmed by displaying, and the position coordinate of the next confirmation defect calculated in the defect inspection step is corrected by the confirmed displacement amount,
A confirmation inspection step of moving and positioning the substrate to be inspected based on the corrected position coordinates of the next confirmation defect to obtain a next confirmation digital image signal and confirm the next confirmation defect. Having a pattern inspection method. 8. A / D by irradiating a substrate to be inspected on which a circuit pattern is formed with charged particles or light, and detecting any of the generated light, secondary electrons, reflected electrons, transmitted electrons, and absorbed electrons.
The detected digital image signal is converted to obtain a detected digital image signal, and the obtained detected digital image signal is compared with a reference digital image signal that can be expected to have originally the same circuit pattern to detect a defect or a defect candidate based on a mismatch. Defect inspection step of calculating the position coordinates of the inspection target substrate, and the inspection target substrate is moved and positioned based on the position coordinates of the confirmation defect calculated in the defect inspection process, and the positioned inspection target substrate When irradiating the upper part with charged particles or light and detecting any of the generated light, secondary electrons, backscattered electrons, transmitted electrons, and absorbed electrons and performing A / D conversion to obtain a digital image signal for confirmation, When an instruction from an operator or a certain condition is satisfied, the reference point image signal is moved to a reference point on the substrate to be inspected, which is registered in advance, and the reference point image signal is detected. The displacement amount of the reference point is calculated by comparing with the image signal, the position coordinate of the checking defect is corrected based on the calculated shift amount of the reference point, and the position coordinate of the corrected checking defect is corrected. The pattern inspection method further comprises a confirmation inspection step of moving and positioning the substrate to be inspected based on the above to obtain a confirmation digital image signal and confirm the confirmation defect. 9. A pattern inspection method for inspecting on the basis of defect information including position information of a defect or defect candidate and image information of the defect or defect candidate on a substrate to be inspected on which a circuit pattern is formed, The amount of deviation between the image information of the defect or defect candidate and the layout information of the substrate to be inspected is calculated, the position information of the defect or defect candidate is corrected by the calculated amount of deviation, and the position of the defect or defect candidate is corrected. A pattern inspection method, wherein information is replaced or added with the position information of the corrected defect or defect candidate. 10. Defect information including position information of a defect or defect candidate on a substrate to be inspected on which a circuit pattern is formed and image information of the defect or defect candidate, and a defect or defect in the received defect information. Detecting the detection image at the defect position by positioning the inspected substrate based on the position information of the candidate, measuring the amount of deviation between the detected detection image and the received defect or defect candidate image information, A pattern defect inspection method, characterized in that the position information of the received defect or defect candidate is corrected with the measured displacement amount, and the defect or defect candidate on the inspection target substrate is confirmed and inspected. 11. Received defect information including image information of a defect position on a substrate to be inspected on which a circuit pattern is formed, detect a detection image at the defect position in the received defect information, and detect the detected image. And the amount of deviation between the received defect position and the image information, and the defect position is corrected by the measured amount of deviation to confirm the defect on the inspected substrate. Defect inspection method. 12. Received defect information including image information of a defect position on a substrate to be inspected having a circuit pattern formed thereon, detecting a detection image at the defect position in the received defect information, and detecting the detected image. A pattern defect inspection method characterized in that a defect on the substrate to be inspected is confirmed by counting the repetition index of the repeating portion based on the above and displaying or adding the counted repetition index to the defect information. 13. A traveling control unit for continuously moving a stage on which a substrate to be inspected is mounted in a predetermined direction, an irradiation optical system for irradiating the substrate to be inspected with charged particles or light, and the irradiation optical system. By irradiating charged particles or light in the system, the light generated from the substrate to be inspected, secondary electrons, reflected electrons, transmitted electrons, or absorbed electrons is detected in synchronization with the traveling of the traveling control unit. An image acquisition unit that obtains a detected digital image signal by A / D conversion and a detected digital image signal obtained by the image acquisition unit are compared with a reference digital image signal that can be expected to have originally the same circuit pattern. And a defect detection unit that detects a defect or a defect candidate based on the mismatch and calculates the position coordinates thereof, and when comparing the detected digital image signal with the reference digital image signal, A pattern inspection apparatus, comprising: a correction unit that calculates a displacement amount and corrects and stores the position coordinates of the defect or defect candidate calculated by the defect detection unit based on the calculated displacement amount. . 14. A traveling control section for continuously traveling in a predetermined direction on a stage having a reference coordinate system on which a substrate to be inspected, on which a plurality of dies are arranged, is mounted, and charged particles on the substrate to be inspected or An irradiation optical system for irradiating light, and light generated by the irradiation optical system for irradiating charged particles or light from the substrate to be inspected according to the array of dies,
An image acquisition unit that obtains a detected digital image signal by detecting any of secondary electrons, reflected electrons, transmitted electrons, and absorbed electrons in synchronization with the traveling of the traveling control unit and performing A / D conversion, and the image acquisition unit. Storage unit that sequentially stores the detected digital image signals that are sequentially obtained by the unit and delays them by the die pitch to sequentially obtain the reference digital image signals that can be expected to have the same pattern, and the detection digital image between the dies or between the cells. The signal and the reference digital image signal are compared to determine the non-coincidence portion as a defect or a defect candidate, and the position coordinates of the defect or the defect candidate are calculated based on the reference coordinate system. The amount of deviation between the detected digital image signal sequentially obtained from the image acquisition unit and the reference digital image signal sequentially obtained from the storage unit is calculated for each divided minute region. The image processing unit to be output, and the deviation amount calculated by the image processing unit for each die in the unit of a minute area is integrated in a unit of the minute area from a desired die to the die in which the defect or the defect candidate is determined. The deviation amount calculation unit that calculates the deviation amount in which the defect candidate is originally located with respect to the reference coordinate system in small area units, and the position coordinates of the defect or the defect candidate calculated by the image processing unit are the deviation amount calculation unit. A pattern inspecting apparatus, comprising: a correction unit that corrects and stores the amount of deviation that is originally located in a unit of a small area in which the defect or defect candidate calculated in step 1 exists. 15. A travel control unit that continuously travels in a predetermined direction on a stage having a reference coordinate system on which a substrate to be inspected, on which a reference die is formed and a plurality of dies are arranged, is previously mounted, An irradiation optical system for irradiating a substrate to be inspected with charged particles or light, and light generated by irradiating the substrate to be inspected with charged particles or light according to a reference die and an array of a plurality of dies of the substrate to be inspected, According to the reference digital image signal of the standard die and the arrangement of the dies, any one of the secondary electrons, the reflected electrons, the transmitted electrons, and the absorbed electrons is detected and A / D converted in synchronization with the traveling of the traveling control unit. An image acquisition section for obtaining a detected digital image signal, a storage section for storing the reference digital image signal of the standard die obtained by the image acquisition section, a detected digital image signal obtained by the image acquisition section, and The non-coincidence portion is compared with the reference digital image signal stored in the storage unit to determine the defect or defect candidate, the position coordinates of the defect or defect candidate are calculated based on the reference coordinate system, and the reference digital image is further calculated. An image processing unit that calculates the amount of deviation of the detected digital image signal with respect to the image signal, and the position coordinates of the defect or defect candidate calculated by the image processing unit are corrected with the amount of deviation calculated in the image processing step. A pattern inspection apparatus comprising: a correction unit that stores the pattern. 16. A travel controller for continuously moving a stage on which a substrate to be inspected is mounted, an irradiation optical system for irradiating the substrate to be inspected with charged particles or light, and charged particles or Any of light, secondary electrons, backscattered electrons, transmitted electrons, and absorbed electrons generated by irradiating light is detected in synchronization with the traveling of the traveling control unit and A /
An image acquisition unit that obtains a detected digital image signal by D conversion and a detected digital image signal obtained by the image acquisition unit are compared with a reference digital image signal that is originally expected to have the same circuit pattern, and they do not match. A defect inspecting section that detects a defect or a defect candidate based on the above and calculates its position coordinates; and a detection digital image signal of a plurality of reference points formed on the inspected substrate obtained by the image acquiring section, Preliminarily registered position coordinates are compared with reference image data of a plurality of reference points to calculate deviation amounts of detected image signals at the respective reference points, and interpolation is performed based on the calculated deviation amounts. By doing so, the deviation correction amount at the position coordinates of the defect or defect candidate calculated by the defect inspection unit is obtained, and the position coordinates of the defect or defect candidate are corrected by the obtained deviation correction amount. Pattern inspection apparatus is characterized in that a correction unit. 17. A storage unit for acquiring defect information including positions of a plurality of defects on a substrate to be inspected and its image information, reading the information from a non-volatile storage medium, or reading and storing the information from an information transfer unit. A stage control unit that positions the inspection target substrate placed based on the defect information stored in the storage unit, and an image detection unit that detects an image signal on the inspection target substrate positioned by the stage control unit. A display unit that displays the detection image detected by the image detection unit and the image information of the defect information obtained from the storage unit in parallel, or by switching or overlapping, and a detection image detected by the image detection unit. And a deviation amount acquisition unit that acquires the deviation amount based on the image information of the defect information obtained from the storage unit. Based on the deviation amount obtained by the deviation amount acquisition unit, the inspection by the stage control unit is performed. It is configured to correct the positioning of the target substrate and detect a detected image of a defect from the image detection unit and display the image on the display unit. Further, in the defect information displayed on the display unit, additional information about the defect A pattern inspection device characterized by being configured so that it can be additionally output. 18. A defect information including positions of a plurality of defects on a substrate to be inspected and image information thereof and positions of reference points and a dictionary image thereof are acquired or read from a non-volatile storage medium or read from an information transfer means. And a storage unit for storing the stored information, a stage control unit for positioning the mounted inspection target substrate based on the defect information stored in the storage unit, and an inspection target substrate positioned by the stage control unit An image detection unit that detects the above image signal, and a display unit that displays the detected image detected by the image detection unit and the image information of the defect information obtained from the storage unit in parallel, or by switching or overlapping. And controlling the stage control unit to move the position of the reference point on the substrate to be inspected and detecting the image signal of the reference point by the image detection unit. The coordinate shift amount is calculated based on the information of the dictionary image stored in the storage unit, and the defect information stored in the storage unit is corrected based on the calculated coordinate shift amount to be inspected by the stage control unit. The position of the target substrate is corrected to detect the image signal of the defect by the image detection unit and can be displayed on the display unit. Further, in the defect information displayed on the display unit, additional information about the defect is added. A pattern inspection apparatus characterized in that it can be additionally output. 19. Defect information including positions of a plurality of defects on a substrate to be inspected and image information thereof and positions of reference points and a dictionary image thereof are acquired or read from a non-volatile storage medium or read from an information transfer means. And a stage control unit for positioning the mounted inspection target substrate based on the defect information stored in the storage unit, and the inspection target substrate positioned by the stage control unit An image detection unit that detects the above image signal, and a display unit that displays the detected image detected by the image detection unit and the image information of the defect information obtained from the storage unit in parallel or by switching or overlapping. And moving the position of the reference point on the substrate to be inspected by controlling the stage control unit to detect the image signal of the reference point by the image detection unit, and the image signal of the detected reference point and the image signal of the reference point. The dictionary image stored in the storage unit is displayed on the display unit in parallel or in a switched or overlapping manner, and the amount of coordinate deviation between the image signal of the reference point and the dictionary image is acquired, and the acquired coordinate deviation is obtained. The defect information stored in the storage unit is corrected based on the amount, the positioning of the substrate to be inspected by the stage control unit is corrected, and the image signal of the defect is detected by the image detection unit to be displayed on the display unit. A pattern inspection apparatus characterized in that it can be displayed, and further can be output by adding additional information about the defect to the defect information displayed on the display section. 20. A storage unit for acquiring defect information including positions of a plurality of defects on a substrate to be inspected and image information thereof, reading the information from a non-volatile storage medium, or reading and storing the information from an information transfer unit. A layout information acquisition unit that acquires layout information on a substrate to be inspected, a stage control unit that positions the substrate to be inspected placed based on the defect information stored in the storage unit, and the stage control unit An image detection unit for detecting an image signal on the substrate to be inspected positioned by, the image information of the layout information acquired by the layout information acquisition unit and the image information of the defect information stored in the storage unit are arranged in parallel. Or a display unit for switching or superimposing and displaying, and acquiring the coordinate deviation amount of both images based on the image information of both images displayed on the display unit. The defect information stored in the storage unit is corrected based on the amount, the positioning of the substrate to be inspected by the stage control unit is corrected, and the image signal of the defect is detected by the image detection unit and displayed on the display unit. A pattern inspection apparatus characterized in that the pattern inspection apparatus is configured to be able to display, and further configured to be able to output by adding additional information about the defect to the defect information displayed on the display unit. 21. A storage unit for acquiring defect information including positions of a plurality of defects on a substrate to be inspected and image information thereof, reading the information from a non-volatile storage medium, or reading and storing the information from an information transfer unit. A layout information acquisition unit that acquires layout information on a substrate to be inspected, a stage control unit that positions the substrate to be inspected placed based on the defect information stored in the storage unit, and the stage control unit An image detection unit for detecting an image signal on the substrate to be inspected positioned by, the image information of the layout information acquired by the layout information acquisition unit and the image information of the defect information stored in the storage unit are arranged in parallel. Or a display unit that displays the images in a switched or overlapping manner, calculates the coordinate shift amount of both image information displayed on the display unit, and calculates the coordinate shift amount based on the calculated coordinate shift amount. The defect information stored in the storage unit is corrected to correct the positioning of the substrate to be inspected by the stage control unit, the image signal of the defect is detected by the image detection unit and can be displayed on the display unit. The defect information displayed on the display,
A pattern inspection apparatus characterized in that it is configured so that additional information regarding a defect can be added and output.