JPH0465616A - Electron beam length measuring method - Google Patents

Abstract

PURPOSE:To easily and accurately extract a feature pattern by enclosing a sample image, which includes a displayed feature pattern, with a rectangular cursor of size corresponding to the magnification of the sample image. CONSTITUTION:A secondary electron generated by a sample 4 through the scanning of an electron beam is detected by a detector 7, whose detection signal is amplified by an amplifier 8 and supplied to an image monitor 9 like a cathode-ray tube. The rectangular cursor 19 is displayed on a scanning electron beam image on the image monitor 9. A cursor instruction circuit 14 is supplied with a signal corresponding to the magnification as a reference signal from a magnification adjusting circuit 12, and consequently the cursor instruction circuit 14 sends an indication to a cursor generating circuit 13 so that a cursor signal for size corresponding to the magnification is supplied from a cursor generating circuit 13 to the image monitor 9. Thus, an operator extracts the feature pattern over a look at the cursor of the size corresponding to the magnification.

Description

[Detailed Description of the Invention] (Industrial Application Field) The present invention scans an electron beam on a sample, and measures the width and length of a specific portion on the sample using signals generated based on this scanning. The present invention relates to an electron beam length measurement method.

(Prior art) In the electron beam length measurement method, the center of the pattern to be measured on the sample is aligned with the optical axis, and then the electron beam is scanned across the pattern to be measured, and this scanning allows detection. The length of the pattern, such as the width, is measured using the secondary electron signals generated.

In this length measurement method, when multiple patterns to be measured are arranged on a specific part of the sample, a length measurement schedule is created in advance from the array information, and the sample stage is moved according to this schedule information. , the centers of the plurality of length measurement target parts are sequentially arranged near the center of the optical axis.

(Problem to be solved by the invention) By the way, the wafer on which the pattern to be measured is formed,
The wafer is placed on a sample stage, and the orientation flat is used as a reference when placing the wafer on the stage. Thereafter, the positions of two or more marks provided on the wafer are detected by an electron beam, and data regarding the rotation and positional deviation of the wafer with respect to the stage is obtained. Using this data, the stage is accurately moved to the actual observation and length measurement point. However, even if the wafer and stage are aligned in this way, there is a limit to the accuracy when placing the wafer on the stage (referred to as orientation flat alignment accuracy) and the accuracy of moving the stage, resulting in an error of about ±8 μm. occurs. Therefore, the center of the part to be measured after the stage is moved and the center of the optical axis do not exactly match, and a positional deviation between the two leads to a decrease in length measurement accuracy.

Therefore, after moving the stage, it is necessary to measure the amount of deviation between the center of the part to be measured and the center of the optical axis by some means, and then move the stage again precisely based on the amount of deviation, or use an auxiliary deflector. It is necessary to move the optical axis of the electron beam to align the center of the part to be measured with the optical axis. If this alignment is performed by moving the optical axis by deflecting the electron beam with an auxiliary deflector,
Since the mechanical factors of the stage are not involved, accuracy is increased.

To measure the amount of deviation mentioned above, the characteristic pattern of the part to be measured is loaded into the computer memory in advance, and then the stage is moved and the pattern of the part to be measured of the sample after the stage stops is loaded into the memory. , for example, by taking a two-dimensional correlation between two types of patterns.

Here, we will discuss problems in extracting feature patterns. FIG. 2 shows the sample surface, and the hatched area is the pattern P. When observing this sample surface at a magnification of 5,000 times (the size of the field of view is 24 μm x 24 μm), if the target object is ideally located at the center of the field of view, the field of view of area a surrounded by the solid line in the figure will be visible on the cathode ray tube. will be displayed. However, as described above, since the positional error of the stage is 8 μm at most, the area included in the field of view will deviate greatly from the ideal area a. The key marks in Fig. 2 indicate the edges of the visual field when the maximum deviation occurs in the visual field, and the key marks 1a to 1d indicate the edges of the visual field when the maximum deviation occurs to the upper left. , when the field of view shifts maximum to the upper right,
Key marks 3a to 3d indicate when the visual field has shifted to the lower left to the maximum, and key marks 4a to 4d indicate when the visual field has shifted to the lower right to the maximum.

The area indicated by the dotted line in this figure is always within the visual field no matter what direction the visual field shifts.

The size of this dotted line area is calculated to be 8 μm×8 μm. From this, extraction of feature patterns requires 5000
At double magnification, the size must be 8 μm x 8 μm or less, but if this size is used, the feature point in the X direction (
Edge) is good because it is included in an area of this size, but
There are no feature points (edges) in the Y direction in this region, and even if the region surrounded by dotted lines is extracted as a feature pattern,
It is not possible to perform alignment in both directions with high precision.

Figure 3 shows the same pattern shown in Figure 2 when observed at a magnification of 4000 times (field of view size is 30 μm x 30 μm).
000x magnification (field size is 40μm x 40μm)
In these figures, the area a surrounded by the solid line or, if the target ideally coincides with the center of the field of view, the area surrounded by the dotted line is the area surrounded by the field of view. This is an area that is always included in the field of view. As is clear from these figures, at a magnification of 4000x, the size of the area that always falls within the field of view is 14 μm x 14 μm.
At a magnification of 3000 times, the size of the area that always falls within the field of view is 24 μm×24 μm. In the case of Fig. 3, when the size of the feature pattern is 14 μm x 14 μm, Y
Since some feature points (edges) in the direction are included, alignment in the Y direction is possible. In the case of FIG. 4, when the size of the feature pattern is 24 μm×24 μm, there are a plurality of feature points (edges) in the Y direction, which is sufficient for alignment in the Y direction.

The present invention has been made in view of the above-mentioned points, and its purpose is to facilitate the extraction of characteristic patterns when pattern matching is used to correct the positional deviation between the measurement target part and the optical axis after the stage is moved. The objective is to realize an electron beam length measurement method that can be used.

(Means for Solving the Problems) The electron beam length measurement method based on the present invention scans a specific area on a specimen sample two-dimensionally with an electron beam, and uses a video signal detected based on this scanning. Electron beam length measurement that searches for a feature pattern extracted in advance in the image, moves this feature pattern to the center of the field of view based on the position information of this feature pattern, and measures the length of the feature pattern. In the method, when extracting a characteristic pattern, a sample image including a characteristic pattern to be measured is displayed, and a rectangular cursor having a size corresponding to a magnification of the sample image is displayed superimposed on the sample image, This method is characterized in that the cursor is positioned on the feature pattern to be measured, and then the area surrounded by the cursor is extracted as the feature pattern.

(Function) When performing pattern matching to find the amount of deviation between the characteristic pattern and the electron beam optical axis, a sample image including the characteristic pattern to be measured is displayed and superimposed on this sample image in order to extract the characteristic pattern. A rectangular cursor whose size corresponds to the magnification of the sample image is displayed, this cursor is positioned over the feature pattern to be measured, and then the area surrounded by the cursor is extracted as the feature pattern.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings. FIG. 1 shows an example of an electron beam length measuring device for implementing the method based on the present invention. 1 is an electron gun, and an electron beam EB from the electron gun 1 is narrowly focused onto a sample 4 by a condenser lens 2 and an objective lens 3. The electron beam EB irradiated onto the sample 4 is transmitted through the deflection coil 5
As a result, a specific area of the sample 4 is scanned by the electron beam. Further, an auxiliary deflection coil 6 is arranged between the deflection coil 5 and the objective lens 3. Secondary electrons generated by irradiating the sample 4 with the electron beam are detected by a secondary electron detector 7, and the detection signal is amplified by an amplifier 8 and supplied to an image monitor 9 and a control circuit 10 such as a computer. .

The deflection coil 5 is supplied with an electron beam scanning signal from a scanning signal generation circuit 11 via a magnification adjustment circuit 12, or this scanning signal is also supplied from the scanning signal generation circuit 11 to an image monitor 9. .

The image monitor 9 is supplied with a cursor signal from the cursor generation circuit 13 as well as the signal from the detector 7 . The cursor generation circuit 13 is configured to generate a cursor signal according to instructions from the cursor instruction circuit 14, and the position of the generated cursor signal on the monitor 9 can be changed by a position shift circuit 15. The cursor instruction circuit 14 is supplied with a signal corresponding to the magnification from the magnification adjustment circuit 12. The auxiliary deflection coil 6 is supplied with a signal from a shift circuit 16, which is controlled by a control circuit 10. A memory 17 is connected to the control circuit 10, and the control circuit 10 also controls a drive mechanism 18 that mechanically moves the sample 4.

Next, the operation of the above-described configuration will be explained.

[J] describes extraction of feature patterns when performing pattern matching. First, the sample 4 is moved so that the pattern to be measured is located directly under the optical axis of the electron beam EB, and the electron beam is scanned two-dimensionally over the sample 4. This scanning is performed by supplying a scanning signal from a scanning signal generation circuit 11 to the deflection coil 5 via a magnification adjustment circuit 12. Secondary electrons generated from the sample 4 based on the scanning of this electron beam are detected by a detector 7, and this detection signal is amplified by an amplifier 8 and then supplied to an image monitor 9 such as a cathode ray tube. . The image monitor 9 is also supplied with a scanning signal from the scanning signal generation circuit 11, and therefore, on the screen from the image monitor 9, a secondary electron image of the sample 7 is set by the magnification adjustment circuit 12. Displayed at magnification.

Next, the cursor instruction circuit 14 is turned on, a cursor signal is generated from the cursor generation circuit 13 according to instructions from the cursor instruction circuit 14, and a rectangular cursor 19 is displayed by superimposing it on the scanning electron beam image on the image monitor 9. let The cursor instruction circuit 14 is supplied with a signal corresponding to the magnification from the magnification adjustment circuit 12 as a reference signal, and as a result,
The cursor instruction circuit 14 instructs the cursor generation circuit 13 to supply a cursor signal having a size corresponding to the magnification from the cursor generation circuit 13 to the image monitor 9. For example, when the magnification is 5000 times, the size of the cursor corresponding to this magnification must be such that it always falls within the field of view due to factors such as stage position error, as discussed in Figure 2. It corresponds to an area of 8 μm×8 μm. Therefore, in the example of FIG. 2, a cursor 19 of a size corresponding to the area indicated by the dotted line in the figure is displayed on the monitor 9. The position of the cursor 19 on the monitor 9 can be arbitrarily moved in both the X and 7 directions by adjusting the potentiometer in the position shift circuit 15.

Now, the operator extracts a feature pattern for pattern matching while observing the image monitor 9. In this embodiment, the operator selects the area indicated by the cursor or the area that can be extracted as a feature pattern. It is composed of Therefore, the position of the cursor 19 displayed on the image monitor 9 is made to match the desired pattern portion by adjusting the position shift circuit 15, and then the control circuit 10
When an instruction is given to extract a characteristic pattern, the secondary electron detection signal in the area surrounded by the cursor 19 is supplied to the memory 17 via the control circuit 10 and stored therein.

Next, the length measurement step after the characteristic pattern is stored in the memory 17 will be explained. If a plurality of length measurement target patterns are arranged in advance on a specific part of the sample, a schedule for length measurement is set from the arrangement information, and the control circuit 10 controls the drive mechanism 18 according to this schedule information.
The sample 4 is moved so that the length measurement pattern is placed on the optical axis. As mentioned above, after this sample movement, the length measurement pattern and the optical axis do not exactly match due to positional errors of the sample stage and the like. Therefore, control circuit 1
0 takes in a secondary electron detection signal based on two-dimensional electron beam scanning at a set magnification after moving the sample, and
Based on pattern matching between the secondary electron detection signal and the characteristic pattern stored in the memory 17, the amount of deviation of the characteristic pattern from the optical axis is determined. A signal corresponding to the determined amount of deviation is supplied to the shift circuit 18, and the shift circuit 18 supplies the auxiliary deflection coil 6 with an electron beam shift signal that aligns the optical axis with the center of the characteristic pattern. . Then, after the optical axis of the electron beam and the center of the characteristic pattern coincide, one-dimensional electron beam scanning is performed to measure the length of the characteristic pattern, and a secondary electron detection signal based on this scanning is sent to the control circuit. The control circuit 10 measures the length of the pattern. After the length measurement of the feature pattern at the specific position is completed, the sample is moved again according to the schedule information, and the length measurement of the feature pattern at the next position is performed in the steps described above.

In this way, feature pattern extraction and actual length measurement are performed. When extracting a feature pattern at a specific magnification, the operator extracts the feature pattern while looking at a cursor whose size corresponds to the magnification. Since it is possible to do
This prevents the extraction of feature patterns that are so large that they are erroneously omitted from the field of view, and enables accurate extraction of feature patterns at a specific magnification. Here, if the magnification is 5000 times, one cursor of the size corresponding to the dotted line area in FIG. As mentioned above, although the feature points of the pattern in the X direction are included, the feature points of the pattern in the Y direction are not included in the area surrounded by the cursor. It is determined that alignment is difficult. In this case, the operator immediately changes the magnification, displays the image at a lower magnification, and also displays a cursor corresponding to that magnification, so that the area surrounded by the cursor is
, Y can include a feature pattern having feature points in both. Thereafter, the data of the characteristic pattern surrounded by the cursor is stored in the memory 18, and pattern matching is performed between this characteristic pattern and the image signal of the same magnification obtained during actual length measurement. It is possible to accurately determine the amount of deviation in both directions.

Although one embodiment of the present invention has been described above, the present invention is not limited to this embodiment. For example, although secondary electrons are detected, reflected electrons may also be detected. Further, although the amount of deviation in the center of the length measurement pattern is corrected by deflecting the electron beam, the optical axis and the center of the measurement pattern may be made equal by moving the stage. Furthermore, we have made it possible to display a cursor according to the magnification, but in that case, we have made it possible to display the cursor at the maximum size according to the magnification.
The size of the cursor may be adjusted to be narrower. Furthermore, although the cursor is moved, the cursor may be fixed in position and the sample image may be moved relative to the cursor to select the characteristic pattern. Wafer alignment involves not only stage accuracy but also orientation flat detection accuracy and wafer chucking accuracy. ) are different, it is preferable to prepare two types of size ranges so that cursors of different sizes can be displayed before and after alignment even at the same magnification.

Alternatively, the standard pattern can be displayed at high magnification over the entire field of view, and the display can be set to the magnification when observing the display and comparing features. In this case, the magnification during pattern matching is not the enlargement magnification but the original magnification.

(Effects of the Invention) As explained above, in the present invention, when performing pattern matching to find the amount of deviation between the characteristic pattern and the electron beam optical axis, when extracting the characteristic pattern, the characteristic pattern to be measured is included. Display the sample image, display a rectangular cursor with a size corresponding to the magnification of the sample image superimposed on this sample image, position this cursor on the feature pattern to be measured, and then move the cursor to Since the enclosed area is extracted as a feature pattern, a feature pattern having feature points in both the X and 7 directions can be easily and accurately extracted no matter what the magnification is.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing an electron beam length measuring device for implementing the method based on the present invention, and FIGS. 2 to 4 are diagrams showing the relationship between magnification and field of view. 1...Electron gun 2...Condenser lens 3
...Objective lens 4・Sample 5...Deflection coil, 6...Auxiliary deflection coil 7
...Detector 8...Amplifier 9...Image monitor 10...Control circuit 11...Scanning signal generation circuit 12...Magnification adjustment circuit 13...Cursor generation circuit 14...Cursor instruction Circuit 15...Position shift circuit 16...Shift circuit 18...Drive mechanism 17...Memory

Claims (1)

[Claims] A specific area on the sample to be measured is scanned two-dimensionally with an electron beam, and a feature pattern extracted in advance is searched for in an image based on a video signal detected based on this scanning.
In an electron beam length measurement method that moves this feature pattern to the center of the field of view based on the position information of this feature pattern and measures the length of the feature pattern, the feature pattern to be measured when extracting the feature pattern. A rectangular cursor of a size corresponding to the magnification of this sample image is displayed superimposed on this sample image, and this cursor is positioned on the feature pattern to be measured, and then,
An electron beam measurement method that extracts the area surrounded by a cursor as a feature pattern.