JPH0375507A - Method and apparatus for inspecting pattern - Google Patents

Abstract

PURPOSE:To make it possible to perform high-magnification inspection at high positioning accuracy by moving a sample stage and a photographing position so that the position of a pattern to be inspected appears at the intended position on a screen. CONSTITUTION:An electron beam emitted from an electron source 1 is projected on a sample 5 through a conversion lens 2, a deflection coil 3 and an objective lens 4. Then, the secondary electrons generated in the surface of the sample 5 are converted into an electric signal in a detector 11. The signal is detected in a secondary-electron detection circuit 14. The result is stored in an image memory 15. The data in the memory 15 are displayed on a CRT display 17 by the control of an electronic computer 16. The expanded image of the surface of a pattern to be inspected is observed. Furthermore, a deflection-signal setting device 12 is actuated by the control of the computer 16. A voltage is applied to the coil 3, and a beam-scanning starting position is freely set. The computer 16 controls driving devices 7 and 8, and a sample stage 6 is two-dimensionnally moved. The intended position of the pattern to be inspected can be observed at the intended position on the display 17.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a pattern inspection method and apparatus for inspecting a high-magnification image of a fine pattern at a desired position, and in particular to a pattern inspection method and apparatus for inspecting a semiconductor pattern without complicated manual operations. The present invention relates to an external appearance inspection method and an inspection device that can automatically inspect the appearance.

[Conventional technology]

A scanning electron microscope (hereinafter referred to as SEM) uses a lens system to narrowly focus the electron beam emitted from an electron gun and scans the electron beam over the sample.The scanning electron microscope (hereinafter referred to as SEM) uses a scintillator or photomultiplier to collect secondary electrons generated from the sample surface. The image is detected by a tube and an image is drawn on a CRT display which is scanned synchronously as a video signal. In this case, the magnification is determined by the ratio between the scanning length of the electron beam on the sample and the corresponding length of the CRT display screen.

SEM is mainly suitable for observing the surface shape of solid samples, and has a characteristic of having a deep depth of focus.

When inspecting a desired position of a pattern to be inspected using (x, y) and an SEM, it is necessary to correctly position the pattern to be inspected so that the desired position is included in the scanning area of the SEM.

Conventionally, a commonly used positioning method involves determining the relationship between the reference position and direction of the SEM scanning area and the reference position and direction of the pattern to be inspected using a specific pattern within the pattern to be inspected prior to inspection. Based on these relationships, a method has been used in which a sample stage on which a pattern to be inspected is placed is moved so that a desired position of the pattern is included in the scanning area of the SEM.

However, with the above method, it is not always possible to obtain high positioning accuracy, so the user manually changes the sample stage position and beam scanning start position every time an inspection is performed.

(x, y) and, for example, as described in Japanese Unexamined Patent Publication No. 62-62207, first, by lowering the magnification of the SEM and widening the scanning area, a desired pattern is formed within the scanning area. A method has been proposed in which a deviation of the pattern reference position is detected from the secondary electron image, the deviation is corrected, the magnification is increased to a high magnification, and the inspection is performed. However, with this method, (a) the desired pattern is drawn in after lowering the magnification for each location, and the magnification is increased again for inspection, so overall adjustment is impossible;
(b) It cannot be used when the positional deviation of the pattern to be inspected is large; and (c) the position of the desired pattern cannot be detected correctly when the pattern to be inspected is complex or when a pattern similar to the desired pattern exists in the vicinity. There were problems such as things that could not be done.

[Problem to be solved by the invention]

In this way, in order to position the desired pattern of the pattern to be inspected at the desired position within the scanning area of the SEM, it is necessary to correctly position the sample stage on which the pattern to be inspected is placed. However, when moving the sample stage, friction and backlash occur, so it is not always possible to move it to the desired position. Furthermore, there is a problem in that the movement of the sample stage is accompanied by distortion, and the distortion changes over time due to the effects of temperature, changes over time, and the like.

(x, y) and SEM, as mentioned above, detects secondary electrons generated from the sample surface using a scintillator or photomultiplier, and scans them synchronously as a video signal.
An image is drawn on the T-display, but when positioning the desired position of the pattern to be inspected at the desired position on the SEM acquired image, distortion associated with SEM beam scanning causes deterioration in positioning accuracy.

Furthermore, the distortion associated with this beam scanning varies depending on the magnification of the SEM and the accelerating voltage.

An object of the present invention is to provide a pattern inspection method and an apparatus therefor that can solve such conventional problems and automatically position a desired position of a pattern to be inspected on a SEM acquired image. There is a particular thing.

[Means to solve the problem]

In order to achieve the above object, the pattern inspection method of the present invention detects the sample stage, detects the position information when moving, and detects the pattern specific position on the image, and the position when moving the imaging position. The information is used to create a coordinate system related to the pattern position, a coordinate system related to the sample stage position, an imaging position movement coordinate system related to the operation amount for moving the imaging position, and an image related to the pixel position of image data arranged two-dimensionally on the image. The relationship between the four coordinate systems of the coordinate system is determined, and from the relationship between the coordinate systems, the pattern position to be inspected given in the pattern coordinate system, the image position to display the pattern position given in the image coordinate system, and the Determine the sample stage position for obtaining the pattern position and image position, and the operation amount for moving the imaging position, respectively, and move the sample stage and imaging position so that the pattern position to be inspected appears at the desired position on the screen. It is characterized by moving position. (x, y) and the pattern inspection apparatus of the present invention includes means for moving the sample stage, means for detecting the position of the sample stage, means for moving the scanning area of a scanning electron microscope, and A means for storing a secondary electron image to be detected as an image and detecting a specific position of the pattern on the image data is provided, so that the desired position of the pattern to be inspected is positioned at the desired position on the image acquired by the scanning electron microscope. Another feature is that the sample stage on which the pattern to be inspected is placed and the scanning area of the scanning electron microscope are moved.

[Function] In the present invention, the sample stage is made movable, the position of the sample stage is also detected, and the scanning area of the SEM is made movable (hereinafter referred to as SEM field of view movement). Further, a secondary electron image generated from the sample by scanning is stored as an image, and the image data is processed to detect a pattern specific position on the image data.

First, as a position calibration, four coordinate systems: pattern coordinate system, sample stage position coordinate system, visual field movement coordinate system, and image data display screen coordinate system are created based on the operation amount given during movement and the observation amount given during detection. Find the relationship between (x, y) and pattern positioning 1. Based on the result of the above calibration, the sample stage position is calculated so that the desired position of the pattern to be inspected is positioned at the desired position on the image, and the sample stage is moved to that position. Furthermore, the amount of positional deviation of the sample table position after movement is determined from the position detection value, and the amount of movement of the SEM field of view corresponding to the amount of deviation is set to perform deviation correction processing.Further, the SEM image is input as an image, and the image data is By detecting the pattern inspection position above, the amount of deviation from the desired position is determined, and the amount of movement of the SEM field of view is corrected. By combining the manipulated variables given during these movements and the observed quantities obtained during detection, the desired position of the pattern to be inspected can be positioned with high precision on the SEM acquired image.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is an overall configuration diagram of a pattern inspection apparatus showing an embodiment of the present invention.

In FIG. 1, 1 is an electron source that generates an electron beam; 2 is an electron source that generates an electron beam;
is a converging lens that converges the electron beam, 3 is a deflection coil for the electron beam, 4 is an objective lens, 5 is a sample with a pattern to be inspected formed on its surface, 6 is a movable sample stage, 7 and 8
9 and 10 are position detection devices for detecting the position of the sample table;
11 is a detector that detects secondary electrons generated by electron lens irradiation, 12 is a deflection signal setting device that sets a deflection reference amount, 13 is a deflection generator that generates a deflection signal for the deflection coil, and 14 is a secondary electron detector The circuit 15 is an image memory 1 that stores a secondary electron signal train according to the synchronization signal of the deflection coil 3.
6 is an electronic computer for setting the deflection reference amount, processing data in the image memory 15, etc., and 17 is a CRT display for displaying data in the image memory 15.

An electron beam emitted from an electron source 1 is converged by a converging lens 2 while traveling, is deflected by a deflection coil 3, and then passes through an objective lens 4 to reach a sample 5. The electron beam scans the surface of the sample 5, and at this time, secondary electrons are generated from the surface of the sample 5. , the secondary electrons are converted into electrical signals by the detector 11, and then detected by the secondary electron detection circuit 14 and stored in the image memory 15. Under the control of the electronic computer 16, the image memory 15
By displaying the data on the CRT display 17, it is possible to observe an enlarged image of the surface of the pattern to be inspected.

Furthermore, under the control of the electronic computer 16, the deflection signal setter 12 is started and a voltage is applied to the deflection coil 3, thereby determining the beam scanning start position (hereinafter referred to as the amount of field of view movement).
can be set freely. (x, y) and the electronic computer 16 can freely move the sample stage 6 two-dimensionally by controlling the drive devices 7 and 8.

As a result, by appropriately setting these two manipulated variables, it is possible to observe the desired position of the pattern to be inspected at the desired position on the display.

In the present invention, these two
Provides a method for setting two manipulated variables.

The pattern inspection apparatus of the present invention includes driving devices 7 and 8 for moving the sample stage, and a sample stage position detection device 9. 10 and a deflection signal setting device 12 for moving the scanning area of the SEM (SEM field of view movement). Deflection generator 13. and a deflection coil 3, and a secondary electron detector 11 for storing a secondary electron image generated from the sample by scanning as an image and detecting a pattern specific position on the image data by processing the image data. Electronic detection circuit 14, image memory 15
and a CRT display 17.

In order to perform highly accurate positioning, we set four coordinate systems: the pattern coordinate system, the sample stage position coordinate system, the field of view movement system, and the image data display screen coordinate system, and calculated the relationship between the coordinate systems to move the moving device. Calculate the amount of operation that should be set. In this alignment procedure, first, as a position calibration, the relationship between these four types of coordinate systems is determined based on the manipulated variables given to these moving devices and the observed quantities obtained from the detector.

In order to position (x, y) and the pattern, calculate the sample stage position based on the above calibration results so that the desired position of the pattern to be inspected is positioned at the desired position on the image, Move the sample stage to that position. Further, the amount of displacement of the sample table position after the movement is determined from the detected value of the position detector, and the amount of movement of the SEM field of view corresponding to the amount of displacement is set to correct the displacement. Furthermore, SEM
By inputting the image and detecting the pattern inspection position on the image data, the amount of deviation from the desired position is determined, and SE
The amount of movement of the M field of view is corrected.

In this way, the amount of operation applied to the sample stage moving device and the SEM field of view moving device is set, and the desired position of the pattern to be inspected is positioned at the desired position on the SEM acquired image.

FIG. 2 is an operation flowchart of position calibration according to the present invention.

In the present invention, the pattern coordinate system CUmiC (U, V), the monitor screen coordinate system "xmi" (x+y), the visual field movement coordinate system CR
Mi”(R,S), and sample stage coordinate system CX=C(X.

We introduce four coordinate systems of Y). In addition, C is co-or
It is an abbreviation of ``dinate''. Here, the pattern coordinate system is a coordinate system in which a pattern is placed on the sample 5 and is necessary for determining the position of a specific pattern to be observed among the patterns. (x, y) and the monitor screen coordinate system are the coordinates required to locate at a specific position on the image when detecting secondary electron radiation due to beam scanning, storing it in memory, and observing it on the image. It is a system. (x,
y) and the visual field movement coordinate system are coordinate systems necessary to keep the viewpoint fixed and move the pattern to be observed to the optical axis. Further, the sample stage coordinate system is a coordinate system necessary for controlling the sample stage while observing the position in order to observe a specific pattern on the sample stage.

Note that the relationship between the sample table coordinates and the motor drive amount, the relationship between the sample table coordinates and the linear encoder detection value, and the relationship between the field of view movement setting value and the actual movement amount are absorbed into the relationship between the coordinate systems described above. Therefore, there is no need to consider it separately. Specifically, the following transformation relationships are used between the four coordinate systems mentioned above.

(1) Relationship between pattern coordinate system and monitor screen coordinate system (ΔR, Δ5) = pattern relative position based on the pattern position at (0, 0), k, 1 to 3s are conversion coefficients.

・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・
・ (1) Here, (ΔU, ΔV) is the pattern relative position with respect to the pattern position at the beam scanning center position (original image origin), Δ represents the error with respect to the reference, and x and y are These are the coordinates of the monitor screen, and bl ~
b1. is the conversion coefficient.

(U) Relationship between visual field movement coordinate system and pattern coordinate system Here, (ΔU, ΔV) is the visual field movement amount before variation.
・ ・ ・ ・ ・ ・ ・ ・ ・ (3) Here,
(U, V) is the pattern coordinate position when the beam vibrates, and is a value without Δ, representing an absolute amount rather than a relative amount. X.

Y is the sample stage coordinates where the sample stage moves, and J II ~ j
□ is a conversion coefficient, and 1 is a constant.

In the relationships (i) and (iji) above, screen distortion and
Distortion due to sample stage movement is taken into account. (x, y) and (ii) are assumed to be linear because the assumption that "image distortion is caused by SEM dynamic distortion and does not depend on the position of the deflection scanning area" generally holds true.

Prior to actual positioning, position calibration is performed to determine the relationship between the coordinate systems described above. Second
The figure shows the procedure for the correction process.

When position calibration is started, first, initial values of each coordinate transformation coefficient are set. In step 201-1, the pattern coordinate system "(U, V)" and the initial value of the number of screens are determined.

Here, Kx is the logical value of deviation amount/setting amount.

Next, in step 201-3, the pattern coordinate system '(U
, V) and the conversion coefficient initial value J of the sample stage coordinate system '(x, y)
, find.

・ ・ ・ ・ ・ (4) Two lines, (LX, Ly) are screen dimensions, (nXl ny)
is the display screen prime number, and M is the magnification.

Next, in step 201-2, the visual field moving coordinate system C(R
, S) and the transformer of the pattern coordinate system C(U, V)...
・ ・ ・ ・ ・ ・ ・ ・ ・ ・ (6) Here, (U,, V,) is the pattern position at the sample stage coordinate origin position.

Next, in step 201-4, steps 201-1 to
201-3, the conversion coefficient initial value B, obtained in step 201-3.

, J, is set.

Next, in step 202, a relationship K between the visual field movement coordinate system and the pattern coordinate system is determined. That is, the pattern feature position coordinate variation amount (ΔUi, ΔVf) with respect to the relative movement amount (ΔR1, ΔSt) of the visual field.

K is obtained by substituting (here, i=1 to N) into each of the following equations.

ΔR to ΔU= ΔU=[ΔU, ΔV]T ΔR=(ΔR1ΔSAT It represents.

Next, in step 203, the relationship B between the pattern coordinate system and the monitor screen coordinate system is determined using the coefficient K described above. That is, the pattern relative position (ΔUi, ΔVi) with respect to the pattern position of the screen coordinate origin with respect to the pattern feature position screen coordinates (xi, yi), (where i=1 to N8
) into the following equations to find relationship B.

ΔU=Bx ΔU=(ΔU, ΔVAT Δx = (x y + X + y 'J
r Next, in step 204, the relationship J between the pattern coordinate system and the sample table coordinate system is determined using the above and B.

That is, the relationship J is determined by substituting the pattern position coordinates (Ui, Vi), (here, i=1 to N) with respect to the sample stage coordinates (Xi, Yi) into the following equations.

U=JX U= (U, V'JT X= (XY,

Each of the above steps will be explained in more detail below.

FIG. 3 is a process flowchart of calibration for determining the relationship between the visual field movement coordinate system and the pattern coordinate system in FIG. 2.

First, in step 301-1, a pattern area to be inspected including a characteristic pattern such as a corner portion is scanned with a beam using a plurality of different visual field movement amounts.

Next, in step 301-2, each visual field movement setting value (ΔR
i, ΔSi), (i=l~N) is stored in the image memory 15 and displayed on the display 17.
Screen coordinates (x r y i )1 (z =
1-N) is detected.

Next, in step 302, the visual field movement setting value (ΔRi,
ΔSt) and detected screen coordinates (x l + y i) (where i=1 to N) are substituted into the following equation to obtain the conversion coefficient C using the calculator 16.

Next, in step 303, the computer 16 calculates the relationship K between the visual field movement coordinate system and the pattern coordinate system using the following equation.

・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・
(8) In this way, the coefficient K is obtained.

FIG. 4 is a flowchart of the procedure for determining the relationship B between the pattern coordinate system and the screen coordinate system in FIG.

The relationship B between the pattern coordinate system and the screen coordinate system can be determined by measuring the screen position and pattern position at multiple points on one screen.

However, to do this, it is necessary to use a pattern with multiple points whose dimensions are already known, and the measurement points must be appropriately distributed within the screen, so the Inspection patterns must be changed. In this embodiment, the field of view is moved at a plurality of points, and the conversion coefficient is obtained from the relationship between the amount of field of view movement and the screen coordinates at that time, thereby making it possible to avoid the above-mentioned problem.

First, in step 401-1, a pattern area to be inspected including a characteristic pattern is scanned with a beam using a plurality of different visual field movement amounts. Next, in step 401-2, the secondary electron image for each visual field movement setting value (ΔRi, ΔSi) (here, 1=1 to N) set in the deflection signal setter 12 by the computer 16 is stored in the image memory 15. Store in. and,
In step 401-3, the screen coordinates (xi, yi) = 1 to N of the pattern feature position are detected by pointing the feature position with a cursor on the display screen of the display 17 or by performing image processing using the computer 16. do.

Next, in step 402, the same calculation as in step 302 is performed, and the field of view movement setting values (ΔRi, ΔSl) and detection screen coordinates (xi, yi) (here, i=1 to N) are substituted into the following equation. Then, the conversion coefficient C is determined by the calculator 16.

Next, in step 403, relationship B is calculated using the following equation using the visual field movement coordinate system and pattern coordinate system conversion coefficient K obtained in step 202.

・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ (
10) FIG. 5 is a process flowchart of calibration for determining the relationship between the pattern coordinate system and the sample stage coordinate system in FIG. 2.

First, in step 501, there are a plurality of pattern positions (sample stage target positions) (Ui) within the pattern to be inspected.

Vi) (here, i=1, 2. . . , N).

Next, perform the following steps for each target value.

That is, in step 502, the sample stage target value (Xi,
Yi) is calculated using the following formula.

X1=(J, IJ,)-'J, lUi ・ ・ ・
・ (II) Kokote, Xi = [X1Yi, Xi, Yi
, 1]7(x, y) and Ui=[Ui, Vila, and [tree]a is the transposed matrix of [*].

Next, in step 503, the sample stage is moved by the drive device 7.8, and the position where the sample stage stops after the movement is completed is detected by the position detection devices 9, 10. Sample stage stop position (Xci
, Yci).

Next, in step 504, the pattern coordinate shift amount (U
Ri, VRi) are calculated using the following formula.

URi=J, (Xc 1-Xi) ・・・・・・ (+
2) Here, URi= [URi, VRi)7Xci=
[*, Xci, Yci, 1]T Note that * is an arbitrary value.

Next, in step 505, the visual field movement amount (ΔR1, ΔSi) for stopping error correction is calculated from the following equation.

ΔRi = 1 URi (13
) Here, ΔRi=(ΔRi, Δ5inT) Next, in step 506, the deflection generator 1 is
3, the above-mentioned visual field movement amount is set, and the stopping error is corrected by the visual field movement.

Next, in step 507, the SEM image is input to the image memory 15 via the secondary electron detector 11, and the computer 16 performs image processing to determine the specified pattern position screen coordinates (XCi, yci) is detected.

Next, in step 508, the deviation from the target screen position (U
Gl, VGi) are determined by the following formula.

UGi=Bxci-Bxi-...-(14) Here, UGi=(UGi, VGi)ax=[xy, x, y, *]a.

Furthermore, in step 509, the (
URi, VRi) and (UGi
, VGi), and stop sample stage position (Xci, Yci).
True pattern coordinates for (U*i, V*1) = (
U-UG i -URi, V-VG 1-VRi) are calculated.

Then, in step 510, the pattern is determined by substituting the stopped sample stage position (Xc t, Yc i) and the true pattern coordinates (u*i, V*i) calculated in step 509 into citizen (3). Find the relationship J between the coordinate system and the sample stage coordinate system.

Note that formula (3) can be rewritten as follows.

・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ (15
) where mi, ni≧O and i1≠i.

1, ll, 12E CL 2+...,N
l(x, y) and the method described above can also accommodate higher-order nonlinearity by extending the relationship between citizen (1) and citizen (3) as shown below.

・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ ・ (
16) Here, mi, ni≧O or i, ≠i.

i, il, i2ε[1, 2,..., NlU′ = [U
' , V']7 x= (xy+xt y1 Figure 6 is a flowchart showing the procedure for positioning a pattern using the relationship between each coordinate system obtained in the above position calibration.

First, in step 601, the position (U, V) to be inspected is
Specify the coordinates of the screen and the screen coordinates (x + y) you want to display.

Next, in step 602, the sample stage movement target coordinates (X, Y) are determined in the following two steps.

In step 6021, pattern coordinates are corrected. That is, using the design coordinate system and the screen coordinate system conversion coefficient B,
Find the pattern position movement amount corresponding to the distance between the screen display position (x+y) and the screen origin (0, 0), and set the pattern position corrected by this amount to the target value (U', V
).

That is, U' = U-B' x・ ・ ・ ・ ・ ・ ・ ・
・ ・ (17) Here, U= [U, Vl r.

Next, in step 6022, a sample stage target value is calculated. The target values (U', V'') are converted into sample stage coordinates using the following equation.

X=(JIJ)-1JIU' ・−・・・−−(+
8) Kokote, X = [XY, X, Y, 11 TUt =
[u/, V']r In step 603, the sample stage is moved by the drive devices 7 and 8 using the sample stage coordinates obtained in step 602 as a target value, and the stop position (Xc, Yc) is determined by the position detection devices 9, Detected at 10.

Next, in step 604, the visual field movement amount (ΔR9ΔS) for correcting the stop position is calculated using the following equation.

ΔR=-1(J'Xc-J'X) ・ ・
・119) Here, ΔR=[ΔR1ΔS]↑ X= [XY, X, Yl A.

xc= [xcYC9XC1Yc]7 Next, in step 606, the SEM image is input to the image memory 15 via the secondary electron detector 11, and the image is processed by the computer 16, thereby determining the specified pattern position on the image. Detect screen coordinates.

Next, in step 607, a second visual field movement correction amount is calculated from the target position (x, y) and the detected position (XC1yc) using the following equation.

ΔR=−1(B' x (, -B' x,) ・
... (20) Here, X = [xy, x, 13th.

X*=CXmVa+xm+y*] Next, in step 605, the visual field movement amount (ΔR2ΔS) is set in the deflection generator 13, and the visual field movement is executed.

Next, in step 608, the correction amount calculated in step 607 is set, and the field of view is moved.

Next, in step 609, the SEM image is input again after the field of view has been moved again.

As described above, in this embodiment, when performing a high-magnification inspection using an SEM, high positioning accuracy can be automatically maintained, and there is no need for complicated manual operations at that time. There is no need to use special patterns for (x, y) and position calibration, so it has a wide range of applicability.

〔Effect of the invention〕

As explained above, according to the present invention, if there is a distortion in the input image, if a nonlinear shift occurs when the sample stage is moved, or if stopping accuracy is insufficient due to the influence of backlash of the sample stage, etc. Even in cases where high-magnification inspection can be performed automatically with high positioning accuracy.

[Brief explanation of drawings]

FIG. 1 is an overall configuration diagram of a pattern inspection device showing an embodiment of the present invention, FIG. 2 is a processing flowchart of position calibration according to the present invention, and FIG. 3 is a visual field movement coordinate system and a pattern coordinate system in FIG. 2. Figure 4 is a flowchart of calibration processing to determine the relationship between the pattern coordinate system in Figure 2 and the monitor display screen coordinate system. Figure 5 is a flowchart of the calibration process to determine the relationship between the pattern coordinate system and the sample in Figure 2. Flowchart of calibration processing for determining the relationship between platform coordinate systems, No. 6
The figure is a flowchart of pattern positioning processing in the present invention. l: Electron source, 2: Converging lens, 3: Deflection coil, 4: Objective lens, 5: Sample, 6: Sample stage, 7, 8: Sample stage driving device, 9, 10: Sample stage position detection device, 11: Secondary electron detector, I2: Deflection signal setter, 13: Deflection signal generator, 14: Secondary electron detection circuit, 15: Image memory, 16:
Electronic computer, 17: CRT display monitor. Figure 1 Figure (Part 2) Figure (Part 1) Figure 4 Figure (Part 1)

Claims (1)

[Claims] 1. In a pattern inspection method in which a pattern on the surface of a sample placed on a sample stage is imaged and the pattern is inspected on the image, a position when detecting and moving the sample stage; Detecting the pattern specific position on the image and using the position information when moving the imaging position, moving the coordinate system regarding the pattern position, the coordinate system regarding the sample stage position, and the imaging position. The relationship between the four coordinate systems of the imaging position movement coordinate system regarding the operation amount for the image data and the image coordinate system regarding the pixel position of the image data two-dimensionally arranged on the image is determined, and from the relationship between the coordinate systems, the above pattern is determined. Move the pattern position to be inspected given in the coordinate system, the image position where the pattern position given in the image coordinate system should be displayed, the sample stage position for obtaining the pattern position and image position, and the imaging position. A pattern inspection method characterized in that the sample stage and the imaging position are moved so that the pattern position to be inspected appears at a desired position on a screen. 2. When determining the relationship between the pattern coordinate system and the sample stage position coordinate system, N pattern positions (Ui, Vi) (here, i = 1, 2, ..., N) are determined in advance on the surface of the sample. ), and for each specified position, use the initial value set in advance as the conversion coefficient between the pattern coordinate system and the sample stage coordinate system,
By converting the pattern coordinates into sample stand position coordinates, the first step of moving the sample stand to the target position, and detecting the position (Xi, Yi) of the sample stand at the end of the movement of the sample stand, the specified pattern is Discrepancies with respect to coordinates (URi, VRi
), and converting the deviation into imaging position movement coordinates,
A second step of moving the imaging position, imaging the pattern to be inspected, detecting the image coordinates of the specified pattern position, determining the deviation from the target image position, and further determining the pattern coordinate values (UGi, The third step is to calculate the difference (URi, VRi) obtained in the second step.
) and the pattern coordinate value (U
Gi, VGi) to determine the sample stage position (Xi,
true pattern design coordinates (U*i, V*i
)≡(U-UGi-URi, V-VGi-VRi) and the fourth step to calculate the sample table position (Xi, Yi) and true pattern design coordinates (U 2. The pattern inspection method according to claim 1, characterized in that the relationship between the pattern coordinate system and the sample stage position coordinate system is determined using the determined data. 3. When moving the sample stage and imaging position so that the pattern position to be inspected appears at the desired position on the screen, the image position of interest ( x, y) and the origin position of the image (0,
0), and calculate the amount of pattern position movement corresponding to the distance between 0) and change the sample stage position (
X, Y) and the imaging position movement coordinates (ΔR_1, ΔS_1) from the error between the sample stage position calculated in the first step and the actual sample stage stop position.
The second step is to image the pattern to be inspected, detect the amount of deviation (Δx, Δy) from the target position on the image, and calculate the imaging position movement coordinates (ΔR_2, Δ
2. The pattern inspection method according to claim 1, wherein the third step of determining S_2) is sequentially executed. 4. When determining the relationship between the pattern coordinate system and the screen coordinate system, consider the distortion at the time of imaging and calculate ▲ mathematical formulas, chemical formulas, tables, etc. for the screen coordinates (x, y) and pattern coordinates (U, V). Yes▼ Here, mi, ni≧0 and i_1≠i_2i, i1,
Define i2∈[1,...,N], and the relationship between the pattern coordinate system and the imaging position moving coordinate system is expressed as ( ▲There are mathematical formulas, chemical formulas, tables, etc.▼, assuming that there is a linear relationship between the two leaps (ΔU, ΔV), and the relationship between the pattern coordinate system and sample stage position coordinate system Therefore, considering the distortion caused by the sample table movement, there are ▲mathematical formulas, chemical formulas, tables, etc. for the part coordinates (U, V) and the sample table movement coordinates (X)▼ where mi, ni≧0 and i_1≠i_2i, ii,
The pattern inspection method according to claim 1, characterized in that i2∈[1,...N]. 5. Image the pattern on the surface of the sample placed on the sample stage,
In a pattern inspection apparatus using a scanning electron microscope that inspects the pattern on an image, a means for moving the sample stage, a means for detecting the position of the sample stage, and a means for moving the scanning area of the scanning electron microscope are provided. and a means for storing a secondary electron image generated from a sample by scanning as an image and detecting a pattern specific position on the image data, so that the desired position of the pattern to be inspected can be detected in the image acquired by the scanning electron microscope. A pattern inspection apparatus characterized by moving a sample stage on which the pattern to be inspected is placed and a scanning area of the scanning electron microscope so that the pattern is positioned at a desired position. 6. The sample is a semiconductor wafer, and when the surface of the semiconductor wafer is irradiated with a charged particle beam according to raster scanning and the reflected electrons or secondary electrons are detected by a detector, the detector is used as an imaging device, and 6. The pattern inspection apparatus according to claim 5, wherein the means for moving the scanning area of the charged particle beam is an imaging position moving means.