JPH11183138A - Method and device for measuring pattern dimension - Google Patents

Abstract

PROBLEM TO BE SOLVED: To scan, in out of alignment, with a predetermined position of a pattern exactly agreed with beam center, without changing beam shape. SOLUTION: A start point for nest scan is shifted in the direction orthogonal to scanning direction each time beam B scanning is complete, for scanning of pattern P at different position, thus, the pattern signal in the scanning direction and that orthogonal to it is detected as 2D distribution. The scan position of a specified signal intensity is calculated from the signal intensity of plurality of pattern signals detected by the plurality of scannings, and a width W or a position C of the pattern P is measured, based on the pattern signal corresponding to the scanning position, so, beam shape is not required to be changed even when displacement in alignment, etc., takes place, and accurate dimension measurement is possible even with low signal intensity of the pattern signal.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for measuring a dimension and a position such as a width of a pattern formed on a substrate surface such as a mask or a reticle by scanning with a beam. It is suitable for measurement on fine patterns.

[0002]

2. Description of the Related Art When measuring the line width and coordinates (position) of a fine pattern formed on a sample surface such as a mask, a reticle, and a wafer for manufacturing a semiconductor, a laser beam is conventionally applied to the sample surface through an optical system. The beam focused on the target is projected and the stage on which the beam or the sample is placed is moved relative to each other before and after the pattern to be measured, so that the beam and the pattern to be measured are relatively scanned. A method of measuring a pattern line width or a coordinate by detecting an edge position of a pattern using scattered light or specularly reflected light generated in the pattern has been used.

As this type of measuring means, for example, Japanese Patent Publication No. 56-25964 discloses a technique for detecting a light beam obtained by diffracting (scattering) a laser beam at a line edge of a pattern, and performing a highly accurate line width measurement. Techniques for doing so have been proposed. With these technologies,
For example, the line width W _{0 of the} pattern to be measured P ₀ as shown in FIG.
In the case of measuring, for example, the beam B projected on the sample surface
The is relatively scanned by moving before and after the measured patterns P _0.

In the measurement of integrated circuit patterns in recent years, the importance of measuring the width W and center coordinates C of a contact hole pattern P as shown in FIG. 7 is increasing. Also,
With the recent high integration, the width W of the contact hole pattern P is also reduced to a size of 1 μm or less. For example, as shown in FIG. 7, when a He—Ne laser beam B is formed on a sample surface, the spot diameter D is at least about 0.8 μm from the optical resolution. On the other hand, since the width W of the contact hole pattern P is also reduced to a size of 1 μm or less, the spot diameter D of the beam B is substantially equal to the size of the pattern P, and in some cases, the pattern P becomes smaller. Therefore, the following problems occur in the dimension measurement.

When measuring a pattern on a sample surface, for example, the XY stage on which the sample is mounted is moved to move the pattern to be measured to the beam position, and the measurement is performed. With respect to the pattern on the sample surface, alignment is performed in the X, Y, and θ directions by an alignment method using image processing or the like using a plurality of typical patterns in advance.

At this time, when trying to measure a fine contact hole pattern P as shown in FIG.
It is necessary to adjust the positional relationship between the pattern and the pattern P with high precision, so that very high-precision alignment is required. Even if very high-precision alignment is performed, if the positional relationship between the pattern used for alignment and the measured contact hole pattern P is deviated from the design value, as shown in FIG. The positional relationship between B and the pattern P is relatively shifted.

On the other hand, the light intensity distribution of the beam B is a Gaussian distribution as shown in FIG. 9, and the light intensity decreases from the center to the periphery of the beam B. When scanning is performed in a state in which the positional relationship is shifted as shown in FIG. 8 (the direction of the arrow is the scanning direction), a portion having a low light intensity around the beam B is used for edge detection. At this time, the intensity of the scattered light signal and the specularly reflected light signal is also reduced. The signal strength can be processed as an electric signal of constant strength by gain control, but if the original signal to be detected becomes smaller,
Naturally, the SN ratio deteriorates, and the measurement accuracy also deteriorates.

When the measurement is performed using the peripheral portion of the beam B, the signal waveform itself changes as compared with the case where the measurement is performed using the central portion, and the width W and the coordinate value of the pattern P are also changed. Will change. By focusing the beam to a smaller diameter, the dependence of the light intensity distribution on the pattern measurement accuracy can be reduced, but the beam focusing limit is limited by the wavelength and the optical system. In particular, there is an absolute limit that the diameter cannot be made smaller than the wavelength of the beam.

Conventionally, in order to solve these problems,
Means for measuring the dimensions of the pattern by changing the shape of the beam have also been used. Specifically, in the above-described example, the measurement is performed by making the beam into an ellipse while the beam has a round shape. That is, in the laser beam transmission system, as shown in FIG. 10, a diffraction slit SL is provided, and the resolution is reduced by diffraction in a direction in which the light beam is narrowed by the diffraction slit SL, and the beam is expanded. As a result, The elliptical beam OV orthogonal to the direction of the diffraction slit SL is projected on the sample surface.

For example, a He-Ne laser (λ = 633n)
When m) is used, the width of the elliptical beam OV is about 0.8 μm, and the length can be freely set to, for example, 3 μm or 5 μm by the width of the diffraction slit SL. Thus, by using the elliptical beam OV, FIG.
As shown in FIG. 1, by making the length of the elliptical beam OV longer than that of the pattern P, the whole area of the pattern P must be within the elliptical beam OV during scanning even if the positional relationship between the elliptical beam OV and the pattern P is slightly shifted. To pass through.

[0011]

However, the above-mentioned conventional dimension measuring means has the following problems. That is, since the elliptical beam OV is a part of the laser beam passed through the diffraction slit SL, there is a problem that the light amount of the beam itself is lower than that of the beam B having a round shape (spot shape). Further, at the time of pattern measurement, since only a part of the elliptical beam OV is projected on the pattern P and contributes to the measurement, the signal intensity of the obtained scattered light or specular reflected light is
It will further decrease. In other words, there has been a problem that the SN ratio of the signal is significantly reduced due to these factors, and the measurement accuracy is deteriorated. Even if the elliptical beam OV is used, the light intensity distribution is still the Gaussian distribution, so that the influence of the relative shift between the scan position and the pattern P is reduced, but the periphery of the end of the elliptical beam OV is reduced. When the light is projected on the pattern P, the problem due to the light intensity distribution cannot be fundamentally solved. Further, when projecting a beam condensed to the wavelength level onto a fine pattern having a size of the wavelength level, a phenomenon occurs in which a part of the beam passes through the fine pattern without being reflected. For this reason, when measuring such a fine pattern, there has been a disadvantage that the reflection efficiency is further reduced.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and has a dimension measuring method capable of accurately measuring a width or a position at a predetermined position of a pattern even when the alignment is shifted without changing the beam shape. It is intended to provide a device.

[0013]

The present invention has the following features to attain the object mentioned above. That is, in the pattern dimension measuring method according to the first aspect, a spot-shaped beam is irradiated on the sample surface on which the pattern is formed, and the beam is scanned relative to the pattern, which is generated by the scanning. In the dimension measuring method for detecting at least one pattern signal of scattered light and reflected light from the pattern and measuring the width or position of the pattern in the scanning direction from the pattern signal, each time the scanning is completed, The pattern is scanned at different positions by shifting the starting point in a direction orthogonal to the scanning direction, and scanning of a predetermined signal intensity is performed based on the signal intensities of a plurality of pattern signals detected by the plurality of scans. A technique of calculating a position and measuring the width or position of the pattern based on a pattern signal corresponding to the scanning position is employed.

In this pattern dimension measuring method, the point at which the next scan is started is shifted in a direction orthogonal to the scanning direction after each alignment is completed, and the pattern is scanned at different positions. Pattern signals in the scanning direction and the orthogonal direction are detected as a two-dimensional distribution. Then, a scanning position of a predetermined signal intensity is calculated from the signal intensities of the plurality of pattern signals detected by the plurality of scans, and the width or position of the pattern is measured based on the pattern signal corresponding to the scanning position. Therefore, it is not necessary to change the beam shape even if an alignment error or the like occurs, and accurate dimension measurement can be performed even if the signal intensity of the pattern signal is small. For example, when measuring the center position of a fine pattern such as a rectangular contact hole pattern and the width at that position, a predetermined signal strength corresponding to the center position of the pattern, that is, the maximum signal strength calculated based on the distribution of the pattern signal Is
The position becomes the center position of the fine pattern in the orthogonal direction and the predetermined position to be scanned. Therefore, the width or position of the pattern is measured based on the pattern signal corresponding to the scanning position at which the maximum signal intensity is obtained.

According to a second aspect of the present invention, there is provided a method for measuring a dimension of a pattern, wherein the beam is condensed in a spot shape with coherent light and irradiated onto the sample surface on which a fine pattern having a wavelength level of the beam is formed. Adopted.

In this pattern dimension measuring method, in particular,
With the fine pattern of the wavelength level of the beam as the measurement target,
Since the beam of the coherent light is condensed in a spot shape and the fine pattern is irradiated by shifting the scanning position a plurality of times, misalignment or transmission of the beam causes low positional accuracy in a single scan and sufficient intensity. Even for a fine pattern in which a pattern signal is difficult to obtain, a scanning position with a predetermined signal intensity can be easily obtained.

According to a third aspect of the present invention, in the method of measuring a pattern dimension, a technique is employed in which the shift amount is a distance smaller than the beam spot diameter.

In this pattern dimension measuring method, since the shift amount is set to a distance smaller than the beam spot diameter, a blank portion between scanning positions in a direction orthogonal to the scanning direction, that is, an area where the beam does not hit. Disappears. That is, by performing scanning at a fine pitch that eliminates the blank portion, more accurate dimension measurement can be performed.

According to a fourth aspect of the present invention, there is provided a pattern dimension measuring method, wherein a technique is employed in which the shift moving direction is controlled by the signal strength of the plurality of pattern signals.

In this pattern dimension measuring method, the shift moving direction is controlled by the signal intensities of a plurality of pattern signals. For example, a position where a predetermined signal intensity is obtained from the signal intensities of the obtained pattern signals is roughly estimated. If the shift moving direction is determined in this way, the scanning position corresponding to the predetermined signal intensity can be obtained more accurately than when the shift moving direction is set to a fixed direction. The amount of shift is also controlled based on the signal intensities of the plurality of pattern signals, so that the number of scans can be reduced, and a scanning position having a predetermined signal intensity can be obtained more accurately and quickly.

In the pattern dimension measuring method according to the fifth aspect, after the scanning position is calculated, the beam is scanned at the scanning position, a pattern signal at the detection position is detected, and the width of the pattern is detected. Alternatively, a technique for measuring the position is adopted.

Further, in the pattern dimension measuring apparatus according to the sixth aspect, as shown in FIG. 12, an irradiation means for irradiating a spot-shaped beam on the sample surface on which the pattern is formed,
Scanning means for scanning the beam relative to the pattern, detection means for detecting at least one pattern signal of scattered light and reflected light from the pattern generated by the scanning, and scanning from the pattern signal A dimension measuring device having a dimension detecting unit for detecting a width or a position of a pattern in a direction, wherein a point at which the next scanning is started is shifted in a direction orthogonal to the scanning direction each time scanning by the scanning unit is completed. A control unit, and a scanning position calculating unit that calculates a scanning position of a predetermined signal intensity by using signal intensities of a plurality of pattern signals detected by the detecting unit by the plurality of scans, and By means,
A technique is employed in which the beam is scanned at the scanning position, and the dimension detection unit detects the width or position of the pattern in the scanning direction from the pattern signal.

In the method and apparatus for measuring the dimension of a pattern, after calculating a scanning position, a beam is scanned at the scanning position, and a pattern signal at the scanning position is detected.
By measuring the width or position of the pattern, it is finally possible to scan the pattern at a predetermined position exactly coincident with the center of the beam.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a dimension measuring method and apparatus according to the present invention will be described below with reference to FIGS. In this figure, reference numeral 1 denotes an XY stage, 2
Is a sample table, 3 is a sample, 4 is an optical system, 5 is He-N
An e-laser light source, 6 is a beam expander, 7 is a bending mirror, and 8 is an objective lens.

As shown in FIG. 2, the dimension measuring apparatus according to the present embodiment is provided on a sample table 2 on an XY stage (scanning means) 1 (a combination of an X stage 1a and a Y stage) movable in the XY directions. Is to measure the line width and coordinates of the fine pattern P formed on the sample 3 such as a mask, a reticle, and a wafer placed on the sample 3. When performing highly accurate coordinate measurement, the position of the XY stage 1 is usually read by the laser interferometer K. In the case where coordinate measurement is not required and only fine dimensions such as line width are measured, reading may be performed by a detecting means having a lower resolution than a laser interferometer such as a linear encoder.

Above the sample 3, an optical system (irradiation means) 4 for detecting the edge of the pattern P is arranged. The optical system 4 expands the laser light (coherent light) L emitted from the He-Ne laser light source 5 by the beam expander 6, reflects the light downward by the bending mirror 7, and then passes through the objective lens 8. It is set so as to converge and project the beam B on the pattern surface of the sample 3. It should be noted that the pattern P to be measured has dimensions at the wavelength level of the laser light L (beam B), and is set to a submicron level similar to the order of the wavelength.

When measuring the width W or the like of the pattern P, for example, to measure the width W in the X direction, FIG.
As shown in (b), the X stage (scanning means) 1a is moved before and after the pattern P to relatively scan the beam B projected on the sample surface. There are two methods for detecting the pattern signal of the pattern P when the beam B is scanned as described above.

One is means for detecting scattered light (diffracted light) from the edge. As shown in FIG. 2, at least one pair of scattered light detectors (detection means) 9 arranged around the objective lens 8 is used. , 9 to detect scattered light. Another detecting means is a method of detecting the specularly reflected light of the beam B. In this detecting means, as shown in FIG. 2, specularly reflected light reflected from the surface of the sample 3 is split by a half mirror 10 and condensed through a condensing lens 11. It is to detect.

The scattered light detectors 9 and 9 and the specularly reflected light detector 12 are electrically connected to a controller (scanning position calculator, dimension detector) 13.
Reference numeral 3 is also electrically connected to the laser interferometer K and the XY stage 1. That is, the control unit 13 controls the XY stage 1 and controls the scattered light detectors 9 and 9.
And a pattern P based on a pattern signal from the regular reflection light detector 12 and a position signal from the laser interferometer K.
Is calculated and calculated.

Next, a method for measuring the size of a fine pattern in this embodiment will be described.

[Multiple Scanning Steps] In the case of measuring a contact hole pattern P as a fine pattern, FIG.
It is assumed that the position of the beam B is shifted from the center of the pattern P due to an alignment error, a pattern position error, and the like as shown in FIG. When scanning is performed in this state (the direction of the arrow is the scanning direction), as described above, the S / N ratio of the scattered light or specularly reflected light deteriorates, and the measurement accuracy decreases.
Since the signal waveform changes as compared with the case where the center position C of the pattern P is scanned, the measured width W and the coordinate value also change.

For this reason, in the present embodiment, FIG.
As shown in (1), the movement of the XY stage 1 (see FIG. 2) controlled by the control unit 13 (see FIG. 1) causes the scan direction (the direction of the arrow indicates the scan direction) with reference to the position 101h where the pattern P is first positioned. (Scan direction) is shifted a scan position (scan position) in a direction orthogonal to direction (scan direction). That is, for example, the positions 101a, 10
Scanning is performed at equal intervals in a fixed direction as 1b, 101c to 101o. That is, the control unit 13 and the XY stage 1
Functions as a scan control unit that shifts the point at which the next scan is started in the direction orthogonal to the scan direction each time the scan is completed.

At this time, the shift amount M of the shift is
Is set to a distance smaller than the spot diameter D of. The scanning position is detected by position measuring means such as a laser interferometer K for reading the XY stage and a linear encoder.

[Scanning Position Calculation Step] At this time, when the light intensity of the peak of the signal of the scattered light or the specular reflection light is observed for each scan, the relationship shown in FIG. 3 is obtained. That is, the light intensity becomes maximum at the scan position 101k where the beam B passes through the center position C of the pattern P. In this way, it is possible to find a position where the scan position passes through the center position C of the pattern P (a scan position with a predetermined signal intensity). The arithmetic processing of these pattern signals is as follows.
This is performed by the control unit 13.

In this embodiment, the actual rescanning position coincides with the center position C of the pattern P. However, if the actual rescanning position does not coincide, the obtained light intensity and the scanning position may be different. The optimum scan position can be determined by interpolation from the relationship. That is, based on the obtained signal of the light intensity, the rescan position can be calculated by using the least squares method or the like.

[Rescanning Step] The pattern P is moved to the optimum scanning position (scanning position having a predetermined signal intensity) obtained in this way, and final measurement is performed. At this time, according to the scattered light detection means described above, the signal intensity of the detected scattered light is detected by the scattered light detectors 9, 9.
Each waveform is obtained as shown in FIGS.

That is, both scattered light detectors 9, 9
By adding the signal intensities (see FIG. 2), a signal as shown in FIG. 4C is obtained. Therefore, the width W in the scanning direction can be obtained by detecting two peak positions of this signal by the control unit 13 and reading the value of the laser interferometer K at that time. Further, the coordinate value of the pattern P can be obtained as the center coordinates of the two detected edges.

On the other hand, according to the detection means using specularly reflected light,
For example, when the pattern P is a chromium pattern formed on a glass surface, the reflectance of chrome is higher than the reflectance of glass, and therefore, as shown in FIG. Signal can be obtained. That is, two edge positions are determined by processing the slope portion of this signal by the control unit 13 (see FIG. 2).
By reading the value of the laser interferometer K at that time, the width W in the scanning direction can be obtained.

It should be noted that the dimension measurement may be performed using at least one of the above two detection means. By using both, more accurate measurement is possible.

In this dimension measuring method, the pattern P is scanned at different positions by shifting the point at which the next scan is started in the direction orthogonal to the scanning direction every time the scanning is completed after the alignment is performed. , A pattern signal in the scanning direction and a direction orthogonal thereto are detected as a two-dimensional distribution. Then, a scanning position at which the maximum signal intensity (predetermined signal intensity) is calculated from the signal intensities of the plurality of pattern signals detected by the plurality of scans, and the pattern P based on the pattern signal corresponding to the scanning position is calculated. Width W
Alternatively, since the center position C is measured, the dimension can be accurately measured even if the signal intensity of the pattern signal is small due to misalignment or the like.

In particular, since the coherent light beam B is focused on the fine pattern P at the wavelength level of the beam B in the form of a spot and the scanning position is shifted a plurality of times, misalignment or misalignment of the beam B is caused. Due to transmission, even in the case of a fine contact hole pattern P in which positional accuracy is low and it is difficult to obtain a pattern signal of sufficient intensity in one scan, a scanning position where the maximum signal intensity is obtained can be easily obtained, and accurate dimensional measurement can be performed. It becomes possible.

Further, since the shift amount M of the shift is set to a distance smaller than the spot diameter D of the beam B, there is no area where the beam B does not hit between the scanning positions in the direction orthogonal to the scanning direction. Accurate dimension measurement becomes possible. Further, the beam B is scanned at the calculated scanning position, the pattern signal at the scanning position is detected, and the width W or the center position C of the pattern P is measured.
Scanning can be performed such that the center position C of the pattern P and the center of the beam B exactly match.

Therefore, in one scan, even if the signal intensity of the pattern signal is low due to misalignment or the like, the center position C of the pattern P is re-scanned by the center of the beam B, and the beam shape is formed by the diffraction slit and the like. Does not need to be deformed, and the dimension is accurately measured with an appropriate signal intensity by the spot beam.

Note that the present invention includes the following embodiments. (1) In the above embodiment, the method in which the beam B is fixed and the XY stage 1 is moved to scan the beam B relative to the pattern P has been described.
A method in which the Y stage is stopped and the beam side is scanned by a movable mirror or the like may be used. In this case, the amount of movement of the movable mirror or the like is read by a laser interferometer or the like in order to detect the amount of movement of the beam.

(2) Although the He-Ne laser light source is used as the beam light source, another light source such as a mercury lamp may be used. (3) A scanning position with a predetermined signal intensity is calculated from the signal intensities of the plurality of pattern signals, and scanning is performed again at this scanning position to measure the line width and position of the pattern. The width or position of the pattern may be obtained by calculating the pattern signal corresponding to the scanning position of the signal intensity of the above from the signal intensity of the plurality of detected pattern signals.

(4) In the above embodiment, the scanning position of the beam is shifted in a certain area in a certain direction and at equal intervals. However, the shift direction and interval of the shift may be changed as appropriate. For example, two scans with a relatively large shift are performed, and then the scan is performed with a smaller shift from the previous scan toward the stronger signal intensity of the pattern signal (closer to the predetermined signal intensity). By feeding back the pattern signal, the moving direction and the interval of the next shift may be calculated.

That is, by controlling the shift moving direction and the interval based on the signal intensities of the plurality of pattern signals, the scanning position having a predetermined signal intensity is obtained as compared with the case where the shift moving direction and the interval are set to be constant. Is required quickly and accurately.

[0048]

According to the present invention, the following effects can be obtained. (1) According to the dimension measuring method of the first aspect, each time scanning is completed, the point at which the next scanning is started is shifted in a direction orthogonal to the scanning direction to scan the pattern at different positions. A scan position of a predetermined signal intensity is calculated from the signal intensities of a plurality of pattern signals detected by a plurality of scans, and the width or position of the pattern is measured based on the pattern signal corresponding to the scan position. Even when an error occurs or when high-precision alignment is not performed, the width and position of the pattern at a predetermined position can be accurately measured without changing the beam shape.

(2) According to the dimension measuring method of the present invention, the beam of the coherent light is focused in a spot shape, and the fine pattern is scanned on the fine pattern of the wavelength level of the beam. The irradiation is performed by shifting the position a plurality of times, so even if it is a fine pattern where the positional accuracy is low and it is difficult to obtain a sufficiently strong pattern signal in one scan due to misalignment or transmission of the beam, a predetermined signal intensity can be easily obtained. Is obtained. Therefore, even when the pattern position on the sample has a large error with respect to the design value, high-precision measurement can be performed on a fine pattern such as a contact hole, and a stable measured value can be obtained. Further, in the alignment of the pattern for measurement, high-precision measurement can be performed on a fine pattern such as a contact hole without performing high-precision alignment, and a stable measured value can be obtained.

(3) According to the dimension measuring method of the third aspect, since the shift amount is set to a distance smaller than the beam spot diameter, the distance between the scanning positions in the direction orthogonal to the scanning direction is reduced. Since there are no untouched areas, more accurate dimensional measurement is possible.

(4) According to the dimension measuring method of the present invention, the shift moving direction is controlled by the signal strength of the plurality of pattern signals, so that the shift moving direction is set to a fixed direction. In addition, the scanning position at which the signal intensity becomes a predetermined value can be obtained quickly and accurately. Therefore, the number of scans can be reduced, and the throughput can be improved.

(5) According to the dimension measuring method of the fifth aspect and the dimension measuring apparatus of the sixth aspect, after calculating the scanning position, the beam is scanned at the scanning position, and the pattern at the scanning position is obtained. By detecting the signal and measuring the width or the position of the pattern, the predetermined position of the pattern and the center of the beam can be accurately matched, and high-precision measurement can be performed. That is, scanning can be performed in a state where the highest signal intensity of the pattern signal can be obtained, and more accurate dimension measurement can be performed.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram showing a positional relationship between a beam and a pattern and a scanning procedure in one embodiment of a pattern dimension measuring method according to the present invention.

FIG. 2 is a schematic layout diagram showing an embodiment of a pattern dimension measuring apparatus according to the present invention.

FIG. 3 is a graph showing a relationship between a beam scanning position and a signal intensity in one embodiment of a pattern dimension measuring method according to the present invention.

FIG. 4 is an explanatory diagram showing a scattered light detection signal in one embodiment of the pattern dimension measuring method according to the present invention.

FIG. 5 is an explanatory diagram showing a specular reflection light detection signal in one embodiment of the pattern dimension measuring method according to the present invention.

FIG. 6 is an explanatory view showing a positional relationship between a beam and a pattern in a conventional example of a pattern dimension measuring method according to the present invention.

FIG. 7 is an explanatory diagram showing a positional relationship between a beam and a contact hole pattern in a conventional example of a pattern dimension measuring method according to the present invention.

FIG. 8 is an explanatory view showing a case where the positional relationship between a beam and a contact hole pattern is relatively shifted in a conventional example of the pattern dimension measuring method according to the present invention.

FIG. 9 is an explanatory view showing a light intensity distribution of a beam in a conventional example of the pattern dimension measuring method according to the present invention.

FIG. 10 is an explanatory view showing a relationship between a diffraction slit and an elliptical beam in a conventional example of a pattern dimension measuring method according to the present invention.

FIG. 11 is an explanatory diagram showing a positional relationship between an elliptical beam and a pattern in a conventional example of the pattern dimension measuring method according to the present invention.

FIG. 12 is a block diagram schematically showing a configuration of a pattern dimension measuring apparatus according to the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 XY stage (scanning means) 3 Sample 4 Optical system (irradiation means) 9 Detector for scattered light (detection means) 12 Detector for regular reflection light (detection means) 13 Control part (scanning position calculation means, dimension detection part) B beam C Center position of pattern D Beam spot diameter L Laser light (coherent light) M Shift movement amount P Contact hole pattern W Pattern width

Claims (6)

[Claims] 1. A spot-shaped beam is radiated on a sample surface on which a pattern is formed, the beam is scanned relative to the pattern, and scattered light and reflected light from the pattern generated by the scanning. In the dimension measuring method of detecting at least one pattern signal of the above and measuring the width or position of the pattern in the scanning direction from the pattern signal, the point of starting the next scan every time the scanning is completed is defined as the point in the scanning direction The pattern is scanned at different positions by shifting in the orthogonal direction, and a scanning position of a predetermined signal intensity is calculated from the signal intensities of a plurality of pattern signals detected by these multiple scans. A dimension measuring method, wherein a width or a position of the pattern is measured based on a corresponding pattern signal. 2. The dimensional measurement according to claim 1, wherein the beam is condensed in a spot shape with coherent light, and is irradiated on the sample surface on which a fine pattern of a wavelength level of the beam is formed. Method. 3. The dimension measuring method according to claim 1, wherein the shift amount is a distance smaller than a spot diameter of the beam. 4. The dimension measuring method according to claim 1, wherein a moving direction of the shift is controlled by signal strengths of the plurality of pattern signals. 5. After calculating the scanning position, the beam is scanned at the scanning position, a pattern signal at the detection position is detected, and the width or position of the pattern is measured. The dimension measuring method according to claim 1, 2, 3, or 4. 6. An irradiation means for irradiating a spot-shaped beam on a sample surface on which a pattern is formed, a scanning means for scanning the beam relatively to the pattern, and A size measuring device having a detecting means for detecting at least one pattern signal of scattered light and reflected light, and a size detecting unit for detecting a width or a position of a pattern in a scanning direction from the pattern signal; A scanning control unit for shifting a point at which the next scanning is started in a direction orthogonal to the scanning direction each time scanning is completed; signal intensities of a plurality of pattern signals detected by the detection unit by the plurality of scannings Scanning position calculating means for calculating a scanning position of a predetermined signal intensity by using, the scanning means, the scanning of the beam at the scanning position ,
A dimension measuring device, wherein the dimension detecting unit detects a width or a position of a pattern in a scanning direction from the pattern signal.