JPH10321497A - Method and device of alignment accuracy measurement - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately work out the amount of deviation between a reference mark and an alignment mark, and to enhance the reliability of the amount of deviation, by a method wherein the error between the reference mark and the alignment mark is computed, and the error is reflected to the positional deviation between the reference mark and the alignment mark. SOLUTION: The value of reference mark is worked out from an image profile by a reference value computer 21 in a control means 11, the above- mentioned value is stored and controlled in a reference value control means 22. The error of alignment mark is worked out from the value of controlled reference mark by an error computing means 23. On the other hand, the positional deviation information of the alignment mark measured from the image profile is computed by a positional deviation computing means 24, and the amount of corrected deviation is computed based on the positional deviation information and the error information by a deviation amount computing means 25. As above- mentioned, the preciseness of alignment can be measured accurately by computing the correct amount of deviation between the reference mark and the alignment mark.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a method and an apparatus for measuring the positioning accuracy of an object to be processed.

[0002]

2. Description of the Related Art For example, when measuring the overlay accuracy of a semiconductor substrate as an object to be processed, the semiconductor substrate is placed on a measurement stage, and a measurement mark provided on the semiconductor substrate is irradiated with light from a light source. Is captured by a CCD (charge coupled device) camera. Then, an image profile is created from the captured image, and the amount of overlay deviation is determined from the image profile.

More specifically, FIGS. 4A and 4B
As shown in FIG. 4C, an image profile 4 as shown in FIG.
Create 3. Then, the coordinates (a, d) of the edge of the reference mark 42 are obtained, and the midpoint XB thereof is obtained. Further, the coordinates (a, d) of the edge of the alignment mark 41 are obtained, and the midpoint XA thereof is obtained. Thereafter, a difference ΔX between the coordinates of XB and XA is obtained as a correction value. The measurement mark includes a reference mark 42 and an alignment mark 41 that matches the reference mark 42.

[0004]

However, when the image of the measurement mark is taken in, the surface of the measurement stage must be moved to the position shown in FIG.
The reference mark Focus surface 45 shown in FIG. In this case, since the position of the reference mark Focus surface 45 is different from the position of the actual alignment mark Focus surface 44,
The image of the alignment mark 41 is erroneously detected as the alignment mark image 46 shifted from the original position due to the inclination C of the optical axis.

Therefore, an error D occurs between the middle point XA of the original alignment mark 41 and the middle point XC affected by the inclination of the optical axis. This error D is a height difference between the alignment mark 41 and the reference mark 42, that is, the alignment mark Fo.
The distance between the cus plane 44 and the reference mark Focus plane 45 is affected. In the current measuring device, the error D is about 5 nm.

On the other hand, with the miniaturization in the semiconductor device manufacturing process, the required specifications of overlay accuracy measurement have been increasing, and it is necessary to improve the overlay accuracy. However, due to the occurrence of the error, the overlay accuracy cannot be accurately measured, the reliability of the shift amount is reduced, and the required specifications cannot be satisfied at present.

The present invention has been made in view of the above points, and provides a positioning accuracy measuring method and apparatus capable of accurately measuring the positioning accuracy and obtaining a highly reliable amount. The purpose is to do.

[0008]

Means for Solving the Problems In order to solve the above problems, the present invention has taken the following means. The present invention is a method for measuring the alignment accuracy of an object using a reference mark and an alignment mark in an optical system using an image profile of the reference mark and the alignment mark. Calculating an error between the reference mark and the alignment mark from the value; calculating a position shift between the reference mark and the alignment mark; and a shift amount based on the error and the position shift. And a step of determining the position.

According to this configuration, the error between the reference mark and the alignment mark is calculated, and this error is reflected on the positional deviation between the reference mark and the alignment mark. Can be accurately calculated. As a result, the reliability of the shift amount is increased,
The positioning accuracy can be accurately measured.

In the method of the present invention, it is preferable that the reference values of the reference mark and the alignment mark are obtained in a state where the signal-to-noise ratio of the image profile is high. It is preferable that the profile is obtained in a state where the symmetry of the profile is high.

The present invention also relates to an apparatus for measuring the positioning accuracy of an object to be processed using an image profile of a reference mark and an alignment mark in an optical system. Movable support means, a light source for irradiating the reference mark and the alignment mark provided on the object to be processed, a means for capturing an image of the reference mark and the alignment mark to create an image profile, and the image profile Control means for obtaining the alignment accuracy from the reference mark, the control means, a reference value calculation means for obtaining a reference value of the reference mark and the alignment mark, and an error between the reference mark and the alignment mark from the reference value Error calculating means for calculating a position shift, and a position shift for calculating a position shift between the reference mark and the alignment mark. Means leaving, to provide an alignment apparatus characterized by having a deviation amount calculating means for calculating a shift amount based on the error and the positional deviation.

According to this configuration, the control means can determine the reference values of the reference mark and the alignment mark, and calculate the error between the reference mark and the alignment mark from the reference values. For this reason, the error can be automatically corrected using this error, and the shift amount between the reference mark and the alignment mark can be accurately calculated.

[0013]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings. FIG. 1 is a schematic view showing an embodiment of the alignment accuracy measuring device of the present invention. In the figure, reference numeral 1 denotes a light source. An optical fiber 2 for guiding light from the light source 1 is connected to the light source 1.

In the optical system of the present apparatus, a CCD camera 3 is installed, and a lens 4, a half mirror 5, and an objective lens 6 are arranged below the CCD camera 3. A measurement stage 9 is provided below the objective lens 6, and a semiconductor substrate 7, which is an object to be processed provided with a measurement mark 8, is provided on the measurement stage 9. Also,
The measuring stage 9 is provided with a driving unit 10 for vertically moving and horizontally moving the measuring stage 9 in directions A and B in the figure. Further, a control unit 11 is connected to the driving unit 10 and the CCD camera 3.

This control means 11, as shown in FIG.
Reference value calculation means 2 for calculating a reference value from an image profile obtained from an image captured by the CCD camera 3
1 and reference value management means 22 for managing the calculated reference value
Error calculating means 23 for calculating an error due to the inclination of the optical axis from the managed reference value, and displacement calculating means 24 for calculating the displacement of the measurement mark 8 from the image profile.
And a shift amount calculating unit 25 for calculating a shift amount from data from the error calculating unit 23 and the position shift calculating unit 24. The control unit 11 also controls the driving unit 10 that moves the measurement stage 9 up and down when calculating the amount of deviation.

Therefore, in the control means 11, the reference value calculation means 21 calculates a reference value from the image profile, stores it in the reference value management means 22 and manages it, and the error calculation means 23 manages it. An error in measurement is calculated from the obtained reference value. On the other hand, the displacement calculating means 24 obtains displacement information of the measurement mark from the image profile, and calculates the displacement amount calculating means 25.
In, a shift amount corrected based on the position shift information and the error information is calculated. Thereby, an accurate shift amount can be obtained.

Next, the operation of the apparatus having the above configuration will be described. First, light from the light source 1 is transmitted through the optical fiber 2
Illuminate the half mirror 5 through the The light path of this light is changed downward by the half mirror 5 and directed toward the substrate. Further, this light passes through the objective lens 6 and irradiates the measurement mark 8 on the substrate 7. Next, the light applied to the measurement mark 8 is applied to the substrate 7 along with the image information of the measurement mark 8.
Then, the light enters the CCD camera 3 through the objective lens 6, the half mirror 5, and the lens 4. The CCD camera 3 captures image information of the measurement mark 8 and creates an image profile of the measurement mark 8 based on the image information.

In the control means 11, specifically, FIG.
The processing is performed according to the flowchart shown in FIG. That is, the measurement stage 9 is moved up and down by the driving means 10 so that the height of the measurement stage 9 is adjusted to the alignment Focus surface of the measurement mark 8 (S1). Next, an image profile is created in that state, and the coordinates of the midpoint XA of the alignment mark are obtained from the image profile by the reference value calculation means 21 as a reference value (S2).

Next, the measurement stage 9 is moved up and down by the driving means 10 so that the height of the measurement stage 9 is adjusted to the reference Focus surface of the measurement mark 8 (S3). Next, the coordinates of the midpoint XC of the alignment mark are obtained as a reference value by the reference value calculation means 21 (S4). Here, the midpoints XA and XC of the alignment mark are managed by the reference value management means 22.

Here, the reference values of the reference mark and the alignment mark are obtained when the signal-to-noise ratio (S / N ratio) of the image profile is high. This makes it possible to accurately obtain a reference value used as a basis for calculating the error. Specifically, the size (interval) of the reference mark and the alignment mark
Is determined, and the signal is matched with the detected signal interval to determine whether the signal is the signal of the alignment mark or the signal of the reference mark.

For example, when the Focus surface is aligned with the alignment mark, the peak of the signal (Signal) of the alignment mark increases, and conversely, the signal other than the alignment mark such as the reference mark (No.
The peak of ise) becomes smaller due to the shift of Focus. On the other hand, when the Focus surface is aligned with the reference mark, the signal of the reference mark (Si
gnal) becomes larger, and conversely, the peak of the signal (Noise) other than the reference mark such as the alignment mark becomes smaller due to the shift of Focus. The S / N ratio is defined as the signal peak value.
The ratio of the peak values of Noise means that the S / N ratio is high.
It means that the value of the / N ratio is large.

Next, the error calculating means 23 obtains the difference D from the midpoints XA and XC of the alignment mark managed by the reference value managing means 22 (S5). Next, the measurement stage 9 is moved to the position of the measurement mark 8 by the driving means 10 (S6), and the measurement stage 9 is further moved up and down by the driving means 10 so that the height of the measurement stage 9 is adjusted to the reference Focus surface of the measurement mark 8. (S7).

In this state, an image profile is created, and the coordinates of the midpoint XD of the reference mark of the measurement mark 8 are obtained from the image profile by the displacement calculator 24 (S8). Further, the position shift calculating means 24 obtains the coordinates of the midpoint XE of the alignment mark of the measurement mark 8 from this image profile (S9).

Next, the shift amount calculating means 25 subtracts the difference D from the midpoint XE of the alignment mark to obtain a corrected midpoint XF (S10), and calculates the corrected midpoint XF by using the difference between XD and XF.
The misalignment amount ΔX is obtained (S11).

According to the above method, the control means obtains the reference values of the reference mark and the alignment mark, and calculates the error between the reference mark and the alignment mark from the reference value. Because it is reflected in
The shift amount between the reference mark and the alignment mark can be accurately calculated. Further, according to the above configuration, it is possible to calculate the error and automatically correct the error to obtain the shift amount.

In the above embodiment, the case where the measurement of the reference value is performed at one point in the error calculation is described. However, the measurement of the reference value may be performed at a plurality of points. In this case, the average value may be used as the error value. Further, in the above embodiment, the case where the reference value is measured on two Focus surfaces has been described. However, the reference value may be obtained at more measurement positions.

In the above embodiment, the case where the reference values of the reference mark and the alignment mark are obtained in a state where the signal-to-noise ratio of the image profile is high is described. The value may be obtained in a state where the symmetry of the image profile is high.

In the above embodiment, the case where the semiconductor substrate is used as the object to be processed is described. However, the present invention can be applied to any object to be subjected to alignment. Further, in the above embodiment, the case where the present invention is applied to the measurement of overlay accuracy of a semiconductor substrate is described. However, the present invention can be applied to measurement of other alignment accuracy.

[0029]

As described above, the alignment accuracy measuring method according to the present invention calculates an error between the reference mark and the alignment mark, and reflects this error in the positional deviation between the reference mark and the alignment mark. Therefore, the shift amount between the reference mark and the alignment mark can be accurately calculated. As a result, the reliability of the shift amount is increased, and the alignment accuracy can be accurately measured.

[Brief description of the drawings]

FIG. 1 is a schematic view showing an embodiment of a positioning accuracy measuring apparatus according to the present invention.

FIG. 2 is a schematic block diagram for explaining control means of the alignment accuracy measuring device of the present invention.

FIG. 3 is a flowchart illustrating an embodiment of a positioning accuracy measuring method according to the present invention.

4A and 4B are diagrams for explaining a conventional alignment accuracy measuring method, in which FIG. 4A is a cross-sectional view showing a measurement mark, and FIG.
Is a plan view showing a measurement mark, and (c) is an image profile of the measurement mark.

5A and 5B are diagrams for explaining a problem in a conventional alignment accuracy measuring method, wherein FIG. 5A is a cross-sectional view showing a measurement mark, FIG. 5B is a plan view showing the measurement mark, and FIG. Image profile.

[Explanation of symbols]

1. Light source, 2. Optical fiber, 3. CCD camera, 4.
Lens 5, half mirror, 6 objective lens, 7 substrate, 8 measurement mark, 9 measurement stage, 10 driving means, 11 control means, 21 reference value calculation means, 22 reference value management means, 23: Error calculating means, 24: Position shift calculating means, 25: Shift amount calculating means.

Claims (4)

[Claims] 1. A method for measuring the alignment accuracy of an object using a reference mark and an alignment mark in an optical system using an image profile of the reference mark and the alignment mark, the method comprising: obtaining a reference value of the reference mark and the alignment mark; Calculating an error between the fiducial mark and the alignment mark from a reference value; calculating a misalignment between the fiducial mark and the alignment mark; misalignment based on the error and the misalignment Determining the quantity. 2. The alignment method according to claim 1, wherein the reference values of the reference mark and the alignment mark are obtained in a state where the signal-to-noise ratio of the image profile is high. 3. The alignment method according to claim 1, wherein the reference values of the reference mark and the alignment mark are obtained with a high symmetry of an image profile. 4. An apparatus for measuring the positioning accuracy of an object using an image profile of a reference mark and an alignment mark in an optical system, wherein the apparatus can move horizontally and vertically while supporting the object. Supporting means, a light source for irradiating light to reference marks and alignment marks provided on the object to be processed, means for capturing an image of the reference marks and alignment marks to create an image profile, and alignment accuracy from the image profile Control means for calculating a reference value of the reference mark and the alignment mark, and an error for calculating an error between the reference mark and the alignment mark from the reference value. Calculating means, a position shift calculating means for calculating a position shift between the reference mark and the alignment mark, A position calculating device for calculating a shift amount based on the error and the position shift.