JP2002050572A - Pattern connection accuracy inspecting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a pattern connection accuracy inspecting method which inspects with high accuracy and at a high speed, the connection accuracy at region boundaries of a pattern exposed or written on many regions by each region. SOLUTION: An image memory 44 stores secondary electronic signals detected by scanning a charged particles beam over each of inspecting regions as inspection images, together with their position coordinates. After fetching of the images of all inspecting regions, an image processor 45 compares the inspecting images with separately prepared reference images to extract corresponding patterns at the inspecting regions to the reference images, thereby detecting the deviation of field connections, etc., from relative positions of these patterns.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for manufacturing a semiconductor device or a liquid crystal panel by sequentially exposing a predetermined area such as a semiconductor wafer, a liquid crystal panel or a mask using an exposure apparatus such as an electron beam drawing apparatus. The present invention relates to a pattern connection accuracy inspection method for inspecting a formed pattern for connection accuracy between exposure regions.

[0002]

2. Description of the Related Art In an electron beam drawing apparatus, an electron beam is deflected in accordance with a drawing pattern to draw a desired pattern. As one of the electron beam writing apparatuses, a variable area electron beam writing apparatus is used. In this apparatus, the cross section of the electron beam is made rectangular by a deflector provided between two rectangular apertures. Is molded.

That is, an image of a first rectangular aperture is projected onto a second rectangular aperture, and an electron beam transmitted through the first rectangular aperture is deflected to change the projection position on the second rectangular aperture. An electron beam having a different cross-sectional area is formed, and the formed electron beam is shot on a material to be drawn.

FIG. 1 shows an example of the variable area type electron beam writing apparatus. Reference numeral 1 denotes an electron gun for generating an electron beam EB, and an electron beam EB generated from the electron gun 1.
Is irradiated onto the first shaping aperture 3 through the irradiation lens 2.

[0005] The aperture image of the first shaping aperture is formed on the second shaping aperture 5 by the shaping lens 4.
The position of the image can be changed by the shaping deflector 6. The image formed by the second shaping aperture 5 is irradiated onto a drawing material 9 via a reduction lens 7 and an objective lens 8. The irradiation position on the drawing material 9 is determined by the positioning deflector 10.
Can be changed by

Reference numeral 11 denotes a control CPU.
Transfers the pattern data from the pattern data memory 12 to the data transfer circuit 13. The pattern data from the data transfer circuit 13 includes a control circuit 14 for controlling the shaping deflector 6, a control circuit 15 for controlling the positioning deflector 10,
A control circuit 16 for controlling the excitation of the objective lens 8, the electron gun 1
Is supplied to a blanking control circuit 18 for controlling a blanker (blanking electrode) 17 for blanking an electron beam generated from the hologram.

A shot time correction memory 19 is connected to the blanking control circuit 18, and a blanking signal from the blanking control circuit 18 is
The correction is made according to the value from the shot time correction memory 19. Further, the control CPU 11 controls the drive circuit 21 of the stage 20 on which the material 9 is placed for moving the material 9 for each field. The operation of such a configuration will now be described.

First, a basic drawing operation will be described. The pattern data stored in the pattern data memory 12 is sequentially read out and supplied to the data transfer circuit 13. Based on the data from the data transfer circuit 13,
The deflection control circuit 14 controls the shaping deflector 6, and the control circuit 15 controls the positioning deflector 10.

As a result, the cross section of the electron beam is shaped into a unit pattern shape by the shaping deflector 6 based on each pattern data, and the unit patterns are sequentially shot on the material 9 to draw a pattern of a desired shape. Done. At this time, the blanking of the electron beam is executed in synchronization with the shot of the electron beam on the material 9 by the blanking signal from the blanking control circuit 18 to the blanker 17.

Further, at the time of drawing in different areas on the material 9, the stage 20 is moved by a predetermined distance according to a command from the control CPU 11 to the stage drive circuit 21. Although not shown, the moving distance of the stage 20 is monitored by a laser measuring device, and the position of the stage is accurately controlled based on the measurement result from the measuring device.

[0011]

Now, even in the above-described electron beam drawing apparatus and other types of apparatuses,
Drawing is performed on a resist on a wafer which is a material to be drawn, and L
When forming an SI pattern, the electron beam deflection range of the device for one die (= chip (10 to 20 mm square)) is as small as about 5 mm square at the maximum. LS
When a pattern having a structure like the gate chain of I is formed, it is necessary to form a pattern connected to the entire chip.

Therefore, to form such a pattern, the stage on which the wafer is mounted and the electric deflection system are controlled so as to connect the pattern every 5 mm. For this reason, in the apparatus shown in FIG. 1, a single deflection system is shown as the positioning deflector 10, but actually, a main deflector for deflection of 5 mm is used as a sub or deflector for a range of 5 mm or less. A sub-sub deflection system is provided. For example, the sub deflection system has a deflection range of 500 μm, and the sub deflection system has a deflection range of 50 μm.

As a writing method other than the variable area type electron beam writing apparatus shown in FIG.
A device incorporating a projection function is also being developed. In this method, an aperture (figure stop) in which several tens of patterns are formed on the electron beam path is provided. The electron beam passing through the aperture is reduced to 1/25 and irradiated on the resist applied to the wafer, and a pattern is drawn in a region (cell) of a maximum of 5 μm square in one shot. Since the shot position can be designated by arbitrary coordinates, it is possible to provide an interval between cells or to make a joint between cells contact each other.
Hereinafter, such a joint is referred to as a “shot connection portion”. With such a configuration, it is possible to freely draw a desired pattern from a small figure of 0.1 μm to a large figure connected to the entire chip.

In a 10 mm square chip drawn by the various electron beam drawing apparatuses described above, the number of field connection portions is 40,000, and the number of connection portions between cells is 4,000,000. This is shown in FIG. In FIG. 2, W is a wafer, and the wafer W
Many chips T are formed therein. Each chip T is virtually divided into a main field F1. Each main field F1 is virtually divided into subfields F2. Subfield F2 is virtually divided into subsubfields F3.

The size of the chip T is, for example, 10 mm ×
The main field F1 has a size of, for example, 5000 μm × 5000 μm. The size of the subfield F2 is, for example, 500 μm × 50.
0 μm, and the size of the sub-subfield F3 is, for example, 50 μm × 50 μm.

As described above, in the electron beam drawing apparatus, the drawing area is virtually divided, and the pattern is drawn for each divided area. That is, after drawing a specific area, the stage is moved or the control of the main deflection system or the sub or sub-sub deflection system is switched to draw a pattern in the writing of an adjacent area. As a result, at a boundary portion between adjacent different regions (hereinafter, sometimes referred to as a “field boundary portion” such as a field boundary portion or a shot connection portion) B, a displacement of a drawn pattern and a spacing error between the patterns are reduced. Occurs.

FIG. 3 shows an example of a pattern shift at the field boundary portion B. The pattern is originally shown in FIG.
What is to be formed as shown in FIG.
The pattern may be shifted in the direction or shifted in the X direction as shown in FIG. 3C. Note that the dotted line in FIG. 3 is the boundary.

The permissible shift in the field boundary of the pattern as shown in FIG. 3 (b) in terms of the performance of the LSI is 1/10 or less of the design size.
With a pattern width of μm, it is 10 nm or less. FIG.
In (c), the connection should be made, but the connection is broken, which is not permissible in actual LSI performance. If the amount of these deviations is equal to or larger than the allowable value, it means that there is an abnormality in the electron beam drawing apparatus, and it is necessary to find this abnormality and readjust the apparatus.

In order to find out which part of the electron beam lithography system has an abnormality, a pattern is drawn on chips on the entire surface of the wafer, and a large number of connection parts in each chip are inspected.
It is necessary to measure the amount and direction of the connection deviation of the pattern at each connection. From the measured result, it is possible to empirically estimate an abnormal portion of the electron beam writing apparatus. If an abnormal spot can be found, it is possible to appropriately perform adjustment and repair of the electron beam writing apparatus.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a pattern capable of inspecting the connection accuracy at a connection portion of a pattern drawn for each of a large number of regions with high accuracy and high speed. The purpose is to realize a connection accuracy inspection method.

[0021]

SUMMARY OF THE INVENTION According to the present invention, each of the regions is sequentially exposed such that a predetermined exposure region of a material to be exposed is exposed, and after the predetermined exposure region is exposed, the other exposure regions are exposed. In the method for inspecting the connection accuracy between the respective exposure regions with respect to the formed exposure pattern, a signal generated from an inspection region including at least two specific patterns existing in different exposure regions is detected, and a detection signal is used. By detecting the position of the specific pattern, by comparing the position of the specific pattern in the reference region obtained from data about the reference region corresponding to the inspection region, and the position of the specific pattern in the inspection region, Since the connection accuracy between the exposure regions in the inspection region is inspected, the connection accuracy at the region boundary of the pattern exposed for each of a large number of regions is inspected with high accuracy and at high speed. Door can be.

[0022]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 4 shows an example of a basic system configuration for implementing the inspection method according to the present invention. In the figure, reference numeral 31 denotes a memory for storing pattern data to be drawn by the electron beam drawing apparatus of FIG. 1, for example. The data in the memory 31 is transmitted to a control processing unit 32 which forms a part of the inspection system.

The control processing unit 32 controls a scanning electron microscope (SEM) included in the inspection system. The SEM includes an electron gun 33 and a condenser lens 3
4. It comprises an objective lens 35, a deflector 36, and a moving stage 38 on which a sample 37 is placed.

The electron beam EB generated and accelerated by the electron gun 33 is narrowly focused on the wafer sample 37 by the condenser lens 34 and the objective lens 35. The irradiation position of the electron beam on the sample 37 is two-dimensionally scanned by the deflector 36. Further, by arbitrarily moving the moving stage 38 in the X and Y directions, the electron beam irradiation area on the sample can be moved. The drive circuit 39 of the deflector 36 and the drive circuit 40 of the moving stage 38 are controlled by the control processing unit 32.

For example, secondary electrons generated by irradiating the sample 37 with the electron beam EB are detected by a secondary electron detector 41. The detection signal of the detector 41 is supplied to the control processing unit 32 via the amplifier 42 and the AD converter 43 as image data representing the sample surface shape. Although a method using secondary electrons will be described here, any signal such as a reflected electron or an absorption current can be used as long as it is a signal obtained from the sample by irradiation of an electron beam.

The image data supplied to the control processing unit 32 is supplied to an image memory 44 and stored. The data stored in the image memory 44 is read and supplied to the image processing unit 45, and the connection accuracy of the pattern is measured based on the image data. The operation of such a configuration will now be described.

First, as the sample 37, a wafer sample in which a pattern is drawn on the entire surface of each chip T by the electron beam drawing apparatus shown in FIG. 1 is used. When a pattern is drawn by the apparatus shown in FIG. 1, for a specific chip Ta (see FIG. 2), a connection portion is not generated with respect to a pattern such as a gate or a contact hole which affects the performance of an LSI device. A pattern drawn by only one deflection system is created and used as a reference chip.

Next, a reference image is collected using the reference chip. Incidentally, since the coordinates of the field connection portion and the shot connection portion of the chip other than the reference chip Ta, that is, the chip to be inspected, are known in advance at the time of drawing, the stage 38 is deflected by the control of the control processing unit 32 based on these coordinates. The detector 36 is controlled to scan an electron beam at the connection-corresponding portion of the reference chip Ta of the wafer sample 37, and as a result, the secondary electron signal detected by the detector 41 is used as an image signal as an amplifier 42 and an AD converter 43. Are taken into the image memory 44 via the control processing unit 32. At this time, the image signal (reference image) of each reference pattern is stored with coordinates.

The photographing magnification of the reference image is the same as the magnification for inspecting another chip to be inspected, or
It is performed at a lower magnification. This is because the raw image has a low magnification (for example, a maximum of 1
This is because, even in the case of / 8), even if the figure is expanded to the inspection magnification by processing such as image interpolation, it has been confirmed by experiments that the property as a reference image is not lost. vice versa,
The reference image may be taken at a larger magnification than during the inspection,
This is not a good idea because the field of view becomes narrow and the number of reference images increases.

When the reference image is captured, if a repetitive pattern having the same shape exists in the visual field, it is not necessary to capture the entire visual field as the reference image. The image may be captured and stored as a reference image. FIG. 5 shows an example of the captured reference image. In this case, an area including three patterns is used as the reference image, but only one of the patterns may be used as the reference image.

The next step is to capture an image of the chip to be inspected. In the inspection system shown in FIG. 4, the images of the chips to be inspected are automatically and sequentially taken in according to the inspection conditions, and stored in the image memory 44 together with the coordinates. That is, the control processing unit 32 controls the drive circuit 40 of the stage 38 in accordance with preset conditions, positions the inspection target area of the wafer sample 37 on the electron beam optical axis, and deflects it via the drive circuit 39. The scanner 36 is controlled to scan the area to be inspected with the electron beam.

The secondary electron signal detected by the scanning of the electron beam in each inspection area is stored in the image memory 44 together with the position coordinates as an inspection image. After the capture of the images of all the inspection areas is completed, the reference image and the inspection image corresponding to the coordinates of the inspection area are stored in the image processing unit 45.
The images are compared with each other to detect a shift in a field connection portion or a shot connection portion. FIG. 5 shows a reference image as described above, and FIG. 6 shows an inspection image obtained from an inspection region corresponding to the reference image of FIG. In FIG. 6, B is a connection area line.

The detection of this shift can be accurately performed by software using various feature extraction algorithms based on advanced techniques. The image of the detected location is marked with a rectangle or a circle, and, for example, the center coordinates of this area are calculated, and at the same time, the shift amount is measured.

When the measurement of all of these deviations is completed, the deviations are displayed in a graph or on a map in order to make it easier to visually grasp the state. The map display includes a wafer map (a map representing the measurement result of one wafer), a chip map (a map representing the measurement result of one chip), a field map (a map representing the measurement result of one field), and the like. In displaying the information, it is easy to understand if the display is performed using a vector graphic representing the magnitude and direction of the deviation. FIG. 7 shows an example of a wafer map performed using vector graphics. In FIG. 7, a solid line represents a vector graphic displayed based on the measurement, and a thin line represents an ideal vector graphic. Here, a line connecting the tips of the vectors is called a vector graphic.

When the wafer map is displayed, the average value of a plurality of shifts on each of the X and Y sides of the chip is vector-displayed, and when the chip map is displayed, the plurality of connection portions of the main field are displayed. If the average value of the deviation is displayed as a vector, the state of the deviation can be grasped more accurately. From such a map, the tendency of the pattern misalignment can be grasped, and from this tendency, the readjustment position of the electron beam lithography apparatus can be determined accurately.

Further, in order to measure the displacement amount of each connection portion with high accuracy, re-measurement is performed for all the inspection areas. In the re-measurement procedure, the wafer sample is moved to the center position of the first inspection area in which the deviation has been detected. Then, the image signal is captured at a high magnification. The magnification at this time is calculated from the amount of deviation measured at the time of detection, and the magnification at the time of re-examination is set so as to be several times larger than the value so as to be suitable for re-measurement.

When all high-magnification images have been captured in this way, the amount of deviation is again measured for each of these images and stored as a database. By adding such a step, the shift amount can be measured with high accuracy. The deviation amount measured with high accuracy is displayed on a map.

By the way, first, a pattern drawn by only one deflection system is formed so as to prevent a connection portion from being generated.
This is used as the reference chip. "However, in the stage of preparation for inspection and measurement, it is not possible to create a special chip in which the critical pattern that affects the performance of the LSI is a pattern that does not include a connection part. It involves complicated work. Therefore, as an alternative, a reference image can be captured as follows.

That is, for example, when there are a plurality of patterns on a wafer, the pattern in a certain region has a connection portion, but the pattern in another region has a connection portion. If not, the idea is to use the pattern having no connection as a reference pattern and the pattern having a connection as an inspection target. In the case of a chip including a connection part, if the same pattern as the pattern of the inspection target area is near the field connection part or in another area of the chip and does not include the connection part, this pattern is stored as a reference image. You can make it. If the same pattern is not in the field of view, the location of the other field is determined by the LSI design CA.
It can be known by comparing and examining the D figure.

The displacement of the connecting portion can be directly detected by image processing as described above. However, a relatively simple method is to calculate the displacement indirectly by image matching. Can be used. That is, in the above-described concept, a pattern having a connection portion is inspected based on a pattern having no connection portion, and a shift amount of the connection portion of the pattern having the connection portion is measured. On the other hand, the following method is a method of measuring a relative value between both of the amounts of displacement of the connection part by comparing the image including the connection part with another image including the connection part with reference to the image including the connection part. .

For example, as shown in FIG. 8A, a certain area P in the visual field is registered as an image of a pattern used as a target when specifying a position in the visual field. Hereinafter, such an image of a pattern used as a target is simply referred to as a “target image”. Such a visual field is referred to as a reference visual field. Note that a region on the sample corresponding to such a reference visual field is a reference region. In addition, this registration may be performed by, for example, selecting and designating an area while observing the scanning image by the operator, or extracting any one from, for example, a repetition pattern in the reference visual field by the image processing technique. Is also good. Image registration is performed between the registered “target image” and an image in the reference visual field (reference image), and an image portion matched in the reference visual field is extracted. The coordinates of the extracted image portion are registered as PO1, PO2, PO3, and PO4.

As described above, in the reference visual field shown in FIG. 8A, there are four patterns having the same shape. For example, if the area P surrounding the upper right pattern is registered as the "target image", the image matching Thus, the coordinates of another pattern having the same shape are automatically registered (of course, the operator may manually perform the registration). Then
Image matching is performed between the “target image” and the inspection image as shown in FIG. 8B to extract a matched image portion. Note that an area on the sample corresponding to such an inspection image is an inspection area. PI of the coordinates of the extracted image part
1, PI2, PI3, and PI4. Further, P
By relatively comparing O1, PO2, and the like with PI1, PI2, and the like, it is possible to calculate the shift amount of each portion. The coordinate positions of PO1, PO2, etc. in the figure are
As long as the coordinates are uniquely determined by the characteristics of each pattern, the coordinates may be, for example, the center position, the upper left corner, or anywhere. In addition, Bx and By in FIG. 8A indicate connection boundaries. Although the drawing pattern in the region P shown in FIG. 8A is completely within the range of the region P, FIG.
In such a case as shown in (1), the region P may be designated and registered so as to cut out a part of a continuous drawing pattern.

The concept of calculating the shift amount is as follows. First, it is assumed that PI1 of the inspection image matches the position of PO1 of the reference image. Based on such an assumption, the deviation of PI2 from PI1 of the inspection image is (PI2-PI1)-(PO2-PO1) = (PI2-
PO2). Here, (PI2-PI1) is a position vector of PI2 viewed from PI1, and (PO2-PO1) is P2
The position vector of PO2 viewed from O1, and (PI2-
PO2) is a shift vector of PI2 based on PO2. Therefore, from this, the drawing area including PI2 is viewed from the drawing area including PI1 (PI2-PO
It can be seen that it is shifted by 2).

In the example of FIGS. 8A and 8B, such a relationship is caused by a shift (PI3-PO3) of the drawing area including PI3 as viewed from the drawing area including PI1, and a shift from the drawing area including PI1. (PI4-PO)
4) etc. can be obtained. Of course, the value of the shift amount calculated according to how the initial assumption is taken differs, so for example, the shift between the left and right areas is the shift of the left area as viewed from the right area, and the shift between the upper and lower areas. The shift may be determined in a unified manner, like the shift of the lower area viewed from the upper area.

With this unification, for example, the deviation of the drawing area including PI4 from the drawing area including PI2 is as follows.
Assuming that the positions of the drawing area including PI2 and the drawing area including PO2 match, (PI4-PI2)-(PO4-PO2) = (PI4-
PO4).

In this way, the relative displacement between two adjacent drawing areas is obtained. In this case, these 2
The two drawing areas need not necessarily be in direct contact with each other, but only need to be within the same field of view for electron beam scanning. This concept is important when considering the deviation between patterns that are not in direct contact. Contact hole C as shown in FIG.
An example is an inspection of a shift between C1 and C2 when another drawing area is sandwiched between a drawing area including 1 and a drawing area including C2.

However, when another drawing area is sandwiched between the drawing area including C1 and the drawing area including C2, a slight field of view (scanning area) is set for the inspection. Caution must be taken. That is, usually, the field of view (scanning area) for the inspection should be set so that a position of some joint is located at the center. Therefore, if any position of the joint Bx on the left side of FIG. 9 is set to be at the center, there is a possibility that C2 on the right side of FIG. 9 may not be in the visual field (scanning area) for inspection. In such a case,
The scanning magnification is reduced to widen the visual field (scanning area) for inspection, or preferably, the center coordinate of the visual field (scanning area) for inspection is set to a position intermediate between the positions C1 and C2 in FIG. It is good to set as follows.

Further, such a "target image" is not always one or one type of pattern in the same reference visual field. A plurality of patterns having different shapes within the same reference visual field may be each set as a “target image”. That is, within the same reference visual field, a region of a pattern of a certain shape A is defined as a first “target image”, a region of a pattern of a different shape B is defined as a second “target image”, and so on. The region of the pattern C is the third “target image”, and the coordinates of each “target image” are detected, and the value of the relative position can be obtained (see FIG. 11).

Further, the PO directly obtained from the image of the reference visual field is used.
1, PO2 ... or (PO2-PO1), (PO4-
As the numerical data such as PO2), data (numerical data) created from CAD data or pattern data may be used. In this way, even if there is a shift in the connection in the reference visual field, there is an advantage that the numerical data used for comparison can use a design value without shift. Further, such numerical data may be handled as attribute data of an image of the reference visual field (reference image) or “target image” (image of the target pattern).

Further, in the above description, it is assumed that the reference visual field includes all “target patterns” in the same visual field, but the reference visual field does not necessarily include all “target patterns” in the same visual field. May be. That is, a plurality of first and second “target images” are extracted from separately acquired images, and their relative intervals ((PO2−PO1), (PO4−PO2), etc.) determined from CAD data or the like. ) May be included in the attribute data. Further, instead of having the attribute data, a plurality of “target images” may be combined to create one reference image.

After doing so, the inspection image obtained from the area to be inspected, PI1, PI2,.
Alternatively, (PI2-PI1), (PI4-PI2), etc. are obtained, and a relative comparison is performed to obtain a deviation.

If only one pattern is present on the chip and includes a connection portion, the following method may be used. A portion of the inspection image that does not include the connection portion of this pattern is cut out as a part of the pattern by image processing, and the positional relationship between the cut out portions is synthesized by using a numerical value of the CAD data to synthesize the image (such a case). Retouch function (retouc
h) Call it a function), and artificially create an image without a disconnection, and use this image as the “target image”.

In producing an image without a connection error, it is possible to know a design pattern of the inspection position by examining a CAD figure corresponding to the coordinates of the inspection position. Can be made.

Further, the acquired image data including the connection portion is transferred to a personal computer, the image is displayed on the computer, and an image is formed by tracing the outline of the image and artificially correcting the connection portion to the correct shape. Alternatively, a black-and-white binary image or a filtered gray level image may be created and transferred to the system of FIG. 4 as a "target image" or reference image.

When a CAD figure is used as a "target image" or a reference image, the CAD pattern figure is downloaded to a dedicated computer, the figure is observed on a CAD monitor, and both sides of the virtual connection are straddled. It is also possible to specify a region including a pattern and download the graphic in this region as a “target image” or a reference image to the inspection system in FIG. Note that C
The AD graphics include those actually reconstructed into what is called pattern data for drawing.

The method of acquiring the reference image, the method of comparing the reference image with the inspection image, measuring the displacement of the connection portion, and displaying the measurement result have been described above. Again here,
The method of acquiring the reference image, the comparison between the reference image and the inspection image, and the measurement of the displacement of the connection portion are collectively described.

The basic concept was as follows. First, a special reference chip with no displacement of the connection portion is prepared, and a reference image is obtained using this reference chip (see FIG. 12A), and an inspection target area having the connection portion (see FIG. 12B).
The amount of displacement of the connection part is measured by comparing with the pattern of FIG. Then, for example, by using such a large number of measured values over the entire surface of the wafer, it is displayed as a distribution map of the amount of deviation over the entire surface of the wafer.

The first method of obtaining a reference image is based on the above-described basic concept. A special reference chip having no displacement of a connection portion is formed, and a reference image is obtained using the reference chip. This is the method (see FIG. 12A). However, there is a practical problem that it is difficult to make a special reference chip without displacement of the connection. Then, the following method was devised.

A second method of obtaining a reference image is to search for a pattern having no connection part by examining CAD data or the like and to find a pattern having no connection part which is the same pattern as the pattern of the inspection target area. This is a method of acquiring an image of a visual field including the reference image as a reference image.

The third method of acquiring the reference image is that even if the pattern in the field of view for acquiring the reference image is a pattern having a connection, the positional relationship between the patterns sandwiching the connection is determined by CAD data or the like. Is used as a reference image indicating a correct positional relationship. According to this method, a plurality of reference patterns acquired from different visual fields can be used as a reference image by determining a positional relationship using CAD data.

A fourth method of acquiring a reference image is a method of acquiring an image of a field of view where there is a possibility that there is a shift in a connection portion as a reference image (see FIG. 12C). Used for measuring the deviation amount.

Next, a method of measuring the displacement of the connection portion will be summarized. The first method of comparing the reference image and the inspection image and measuring the displacement of the connection portion is to compare the reference image (FIG. 12A) without the displacement of the connection portion and the inspection image (FIG. 12B). This is a method of measuring the amount of displacement of the connection portion.

The second method of comparing the reference image and the inspection image and measuring the displacement of the connection portion is as follows. The reference image (FIG. 12C) and the inspection image (FIG.
This is a method of measuring the relative shift amount of the connection portion by comparing (b)).

The reason why such a method of measuring the relative shift amount is useful is that the measurement result does not concern the shift of the connection portion of each pattern, but rather the entire drawn (whole wafer or chip). This is because there is a case where the purpose is to know the distribution state of the shift amount of the connection portion in the whole. For example, if the deviation of the same pattern as the pattern included in the reference image is measured over the entire wafer using one reference image, even if there is a deviation in the reference image, the difference between the respective parts of the wafer is sufficiently different. Can be grasped.

Although the embodiment of the present invention has been described above, the present invention is not limited to this embodiment, and other modifications are possible. For example, the connection gap is detected and measured after capturing all the images of the inspection location, but the connection gap may be detected and measured every time an image of each inspection location is acquired. .

Further, when a reference pattern having no connection portion is formed for a specific chip of a wafer, it is not always necessary to form a specific chip for all wafers of the same kind of wafer.

Further, the object to be inspected is not limited to a wafer drawn by a direct electron beam drawing apparatus, but is prepared by an electron beam drawing apparatus, a laser drawing apparatus, or a light and electron mix and match system. Inspection / measurement of an LSI pattern exposure mask can be performed in the same manner. Also, a daughter mask made by connecting patterns formed by projecting and exposing a master mask to 1/5 reduction by a light exposure method, using a mask formed by a stepper exposure apparatus using light, an electron beam, UV, or EUV as a light source. Inspection and measurement can be performed in the same manner.

An exposure mask is a pattern in which a pattern is drawn on a thick quartz glass, and the inspection step may be a step in which the material of the pattern remains a resist or a step in which a mask manufacturing process has advanced. . Furthermore, a stencil mask or an X-ray mask using a silicon wafer as a base material instead of glass can also be inspected.

Further, in addition, in the above-described embodiment, the pattern formed by scanning with an electron beam or “drawing” by an electron beam having a variable shape is mainly described. However, according to the inspection method of the present invention, even a pattern in which predetermined areas are successively "exposed" and connected by a collective exposure method in which predetermined areas are exposed at once with light or an electron beam / ion beam or a laser. It is obvious that the present invention can be applied to a pattern having a joint.
In this sense, “exposure” generally has a broader meaning than “drawing”, but here, both terms are to be interpreted in a broad sense that includes both.

In the above-described embodiment, an electron beam inspection apparatus has been described as an example of a charged particle beam inspection apparatus. However, the present invention is not limited to this, and other types of beam inspection apparatuses, for example, a laser The present invention can be similarly applied to a beam or ion beam inspection apparatus or an optical inspection apparatus.

Further, in the above-described embodiment, a method of acquiring an image by a scanning method has been described. However, for example, a method of acquiring an image projected / enlarged by an electron or light using a CCD camera or the like may be used. Needless to say, such an inspection can be performed.

Further, in the above-described embodiment, the position data and the like are attached to the image data and the like for convenience of explanation. Conversely, the image data and the like are attached to the position data and the like. You may.

[0073]

As described above in detail, according to the present invention, by comparing data acquired from an inspection area with data acquired from a reference area, each of a plurality of exposure areas in the inspection area is compared. The connection accuracy at the boundary of the exposed area of the exposed pattern can be inspected at high speed with high accuracy. At this time, in the actual sample for acquiring the data on the reference region, even if the pattern has a deviation at the connection portion, the CAD data is used, and the like.
This can be equivalent to data obtained from a pattern having no displacement at the connection portion. Furthermore, by relatively comparing the data acquired from the inspection area with the data acquired from the reference area, even if there is a deviation in the data on the reference area at the connection part, the predetermined purpose can be achieved. .

[0074]

[Brief description of the drawings]

FIG. 1 is a diagram showing a variable area type electron beam writing apparatus.

FIG. 2 is a diagram schematically showing a relationship between a wafer and chips and fields formed on the wafer.

FIG. 3 is a diagram showing a pattern shift at a field boundary.

FIG. 4 is a diagram showing an example of a basic system configuration for implementing an inspection method according to the present invention.

FIG. 5 is a diagram illustrating an example of a captured reference image.

FIG. 6 is a diagram illustrating an example of an inspection image.

FIG. 7 is a diagram illustrating an example of a wafer map in which a deviation of connection accuracy is displayed by a vector graphic.

FIG. 8 is a diagram for explaining image matching processing in the case of a repeated pattern.

FIG. 9 is a diagram for explaining a measurement process of a positional shift of a contact hole.

FIG. 10 is a diagram for explaining another example of the designated range of the reference image.

FIG. 11 is a diagram for describing an example in which a reference image has a different shape.

FIG. 12 shows a reference image having no connection portion, an inspection image,
It is a figure for explaining an example with a reference picture which has a connection part.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 31 ... Data memory 32 ... Control processing unit 33 ... Electron gun 34 ... Condenser lens 35 ... Objective lens 36 ... Deflector 37 ... Wafer sample 38 ... Moving stage 39 ... Deflector drive circuit 40 ... Stage drive circuit 41 ... Secondary electron detection Unit 42 Amplifier 43 AD converter 44 Image memory 45 Image processing unit

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G01B 11/24 F K F Term (Reference) 2F065 AA03 AA07 AA09 AA12 AA14 AA56 BB02 BB03 BB27 CC19 DD03 DD06 FF04 FF26 FF61 GG01 GG04 MM16 PP22 PP24 QQ24 QQ25 QQ39 SS02 2F067 AA03 AA13 BB02 BB04 BB21 CC17 EE04 EE10 GG06 GG08 HH06 HH13 JJ05 KK04 LL16 RR24 RR30 RR35 5F056 AA04 BA08 BB01 CA09

Claims (12)

[Claims] 1. Exposure in a predetermined exposure area of a material to be exposed,
In the method of inspecting the exposure pattern formed by exposing each area in sequence so as to expose the other exposure areas after the exposure of the predetermined exposure area, the connection accuracy between the respective exposure areas is examined. A signal generated from an inspection area including at least two specific patterns to be detected is detected, the position of the specific pattern is detected from the detection signal, and the position of the specific pattern is determined from data on a reference area corresponding to the inspection area. A pattern connection accuracy inspection method for inspecting the connection accuracy between exposure regions in the inspection region by comparing the position of a specific pattern in a reference region with the position of the specific pattern in the inspection region. 2. The pattern according to claim 1, wherein the position of the specific pattern in the inspection area or the reference area detects a relative position of another specific pattern with respect to an arbitrary specific pattern in the area. Connection accuracy inspection method. 3. The comparison is made by comparing a relative position of another specific pattern in the inspection area with respect to an arbitrary specific pattern in the inspection area, and the reference area corresponding to the relative position of the specific pattern in the reference area. 2. The pattern connection accuracy inspection method according to claim 1, wherein the difference is obtained as a difference from a relative position of another specific pattern. 4. Any of the specific patterns in the inspection area has a different pattern shape from other specific patterns in the inspection area, and the specific pattern of the reference area in the reference area corresponds to this. The pattern connection accuracy inspection method according to any one of claims 1 to 3, wherein any one has a pattern shape different from other specific patterns in the reference area. 5. The data for the reference area is obtained by detecting a signal generated from an exposure pattern formed so as not to have a connection portion to obtain an image.
The pattern connection accuracy inspection method according to any one of the above. 6. The data processing apparatus according to claim 1, wherein at least one or a part of the data of the reference area or the data of the position of the specific pattern is created from CAD data or pattern data. Pattern connection accuracy inspection method. 7. The pattern according to claim 1, wherein the data of the position of the inspection area or the data of the position of the reference area is accompanied by image data or data extracted from the image data. Connection accuracy inspection method. 8. The pattern connection accuracy inspection method according to claim 1, wherein the specific pattern image data is accompanied by position data or data extracted from the image data. 9. The method according to claim 1, wherein the data of the position of the inspection area or the data of the position of the reference area is accompanied by data extracted or created from CAD data or pattern data. The described pattern connection accuracy inspection method. 10. The image data of the specific pattern includes:
7. The pattern connection accuracy inspection method according to claim 1, wherein data extracted from or created from position data, CAD data or pattern data is attached. 11. The pattern connection accuracy inspection method according to claim 1, wherein the signal is generated by irradiating the inspection area or the reference area while scanning the focused electron beam. 12. The signal is generated by irradiating a light or electron beam on the inspection area or the reference area, and detecting the signal includes projecting light or electrons from the inspection area or the reference area. The pattern connection accuracy inspection method according to claim 1, wherein the pattern connection accuracy inspection method is configured to detect an enlarged image.