JPH0658730A - Measuring method of precision of superposition - Google Patents

Abstract

PURPOSE:To enable measurement of the precision of superposition by a measuring machine by using patterns for visual inspection, by a method wherein only first and second discrete pattern groups are scanned by a light beam, edge positions of the patterns are determined by sensing reflected lights therefrom and the precision of superposition of layers of the first and second patterns is determined on the basis of the edge positions. CONSTITUTION:A laser light flux of a light source 1 is scanned by a reciprocating mirror 14, shifted in the direction perpendicular to the scanning direction by a parallel flat plate 15, transmitted through a beam splitter 3, reflected by a total reflection mirror 4 and cast on an objective lens 5. The light converged by the lens 5 is imaged on a sample 7 and made to scan. A reflected light from the sample 7 enters a reflected light detector 8. A signal obtained through photoelectric conversion by the detector 8 is converted into a digital signal and then taken in a memory 12 synchronously with an interference signal of an interferometer 10. Data in the memory 12 are processed by CPU 13 so as to determine edge positions, the precision of superposition is calculated and a measured value is displayed in CRT and the like.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an overlay accuracy measuring method such as a lithography process, and more particularly to an overlay accuracy measuring method for optically measuring overlay accuracy.

[0002]

2. Description of the Related Art FIG. 4 shows a pattern called a vernier for visually measuring overlay accuracy. The first pattern layer has a square pattern A ₁ ... A
_N is formed, and the pitch of the patterns A ₁ ... A _N (distance between the centers of the patterns) is set to P 1. a1, b
1, a2, b2 ... aN, bN (normally N is an odd number) indicate the edges of the patterns A ₁ ... A _N. The second pattern layer rectangular pattern _B 1 ... _{B N} are formed, the pattern _{B 1} ... B
The pitch of _N (the distance between the centers of the patterns) is set to P2. c1, d1, c2, d2 ... cN, dN indicate the edges of the patterns B ₁ ... B _N. Pitch P1 and Pitch P
Since the 2 slightly different, as shown in FIG. 4, the position of the pattern B ₁ ... B _N overlapping the pattern A ₁ ... A _N is shifted for each set.

When the overlay accuracy is 0, the pattern A ((N + 1) / at the center of the pattern row of the first pattern layer is used.
2), that is, the center of the (N + 1) / 2th pattern and the pattern B ((N
+1) / 2), that is, the center of the (N + 1) / 2th pattern is designed to coincide with each other. In this case, when the entire vernier is visually observed, the pattern sequence of the first pattern layer Pattern A ((N + 1) / 2) and the second pattern in the center of
Pattern B ((N +
The center of 1) / 2) appears to overlap.

When the overlay accuracy is measured by visual inspection, it is read at what position the centers of the patterns of the first pattern layer and the second pattern layer overlap. When overlapping in the m-th pattern, the overlay accuracy R is R = [(P1-P2) * {m- (N + 1) / 2}] / 2.

On the other hand, when measuring the overlay accuracy with a measuring machine, a vernier is not used and a pattern dedicated to the measuring machine is formed separately. 5A and 5B are top views showing a measurement pattern dedicated to the measuring machine. The edges of the patterns C and E formed in the first pattern layer are located at points a and b on the coordinate axis with an arbitrary point as 0, and the second at points c and d.
The edges of the patterns D and F formed on the pattern layer are located.

The center of the pattern C of the first pattern layer and the center of the pattern D of the second pattern layer in FIG. 5A are designed to coincide with each other. The center of the pattern E of the first pattern layer and the center of the pattern F of the second pattern layer in FIG. 5B are designed to be displaced by the width W.
The overlay accuracy R is expressed by the following equation.

In the case of FIG. 5A: R = {(a + b)-
(C + d)} / 2 In the case of FIG. 5B: R = {(c + d)-(a + b)} /
2-W Normally, when the measurement is performed using the overlay accuracy measuring device, the edge positions a, b, c, d in FIGS. 5A and 5B are used.
Then, the overlay accuracy R is calculated by a measuring machine.

[0008]

However, since the pattern used for the visual measurement and the pattern used for the measurement by the measuring machine are different as described above, it is difficult to calibrate the visually measured value and the measured value by the measuring machine. There was a problem.

The present invention has been made in view of the above circumstances, and an object thereof is to provide an overlay accuracy measuring method capable of measuring overlay accuracy by a measuring machine using a pattern for visual inspection. Is.

[0010]

In order to solve the above problems, the overlay accuracy measuring method of the present invention is a first pattern in which a first pattern having a width equal to that of a first pattern layer is formed at a first pitch. And a second pattern having a width equal to that of a second pattern layer adjacent to the first pattern layer at a second pitch slightly different from the first pitch, and the first pattern and its arrangement. A vernier-type overlay accuracy measuring pattern formed of a second pattern group formed so as to partially overlap in a direction orthogonal to the direction is used to form the first pattern layer and the second pattern layer. In the method of measuring overlay accuracy, a first step of scanning only the first pattern group with a light beam and receiving reflected light thereof to obtain an edge position of the first pattern; A second moving to a direction orthogonal to the array direction of the first pattern, scanning only the second pattern group with the light beam, receiving the reflected light, and obtaining an edge position of the second pattern; Step, and the superposition of the first pattern layer and the second pattern layer according to the edge positions of the first pattern and the second pattern obtained in the first step and the second step. Ask for accuracy.

[0011]

Since the measurement accuracy of the overlay can be measured by the measuring machine using the measurement pattern for the visual inspection, the visual calibration of the measured value becomes easy.

[0012]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 2 is a block diagram of a measuring machine for carrying out the overlay accuracy measuring method according to an embodiment of the present invention.

A reciprocating mirror 14 is arranged on the optical path of the laser light output from the light source 1 such as a laser oscillator, and a parallel plate glass 15 and a condenser lens 2 are arranged on the optical path of the laser light reflected by the reciprocating mirror 14. A beam splitter 3 and a total reflection mirror 4 are arranged.

An objective lens 5 and a stage 6 are arranged on the optical path of the laser light reflected by the total reflection mirror 4, and the stage 6
A sample 7 is placed on the top.

A reflected light detector 8 is arranged on the optical path of the reflected light reflected by the sample 7 and reflected by the beam splitter 3.

An interferometer 10 and a control circuit 9 are connected to the reciprocating mirror 14. The interferometer 10 and the reflected light detector 8 convert the analog signal into a digital signal A /
The memory 12 is connected to the D converter 11, the A / D converter 11 is connected to the memory 12, and the memory 12 is connected to the CPU 13.

The laser beam emitted from the light source 1 is expanded by a beam expander (not shown), scanned by the reciprocating mirror 14, moved by the parallel plate glass 15 in a direction perpendicular to the scanning direction, and a condenser lens. 2 and the beam splitter 3, the light is reflected by the total reflection mirror 4 and enters the objective lens 5, is condensed by the objective lens 5, is imaged in a spot shape on the sample 7, and is scanned. The reflected light reflected by the sample 7 is condensed by the objective lens 5, is reflected by the total reflection mirror 4 and the beam splitter 3, and enters the reflected light detector 8.

The position of the reciprocating mirror 14 is controlled by the interferometer 10 and the control circuit 9, and the sample 7 is scanned by the spot light. The rotation angle of the parallel plate glass 15 is controlled by the control circuit 9 about an axis perpendicular to the plane of the drawing,
It is possible to move the spot light on the sample 7 by an arbitrary distance in a direction perpendicular to the scanning direction of the reciprocating mirror 14. The signal photoelectrically converted by the reflected light detector 8 is converted into a digital signal by the A / D converter 11 and sequentially captured in the memory 12 in synchronization with the interferometer signal of the interferometer 10. The interferometer 10 also controls the reciprocating mirror 14, and the address of the memory 12 into which the data is captured and the image forming position on the sample 7 are uniquely determined. That is, the interval between data at one address and the next address in the memory 12 corresponds to the interval of interferometer pulses. The data taken into the memory 12 is processed by the CPU 13 and the position of the edge described later is determined.

FIG. 1 is a top view showing a pattern for measuring overlay accuracy, which is the same as the pattern (vernier) for visually measuring overlay accuracy. Patterns G ₁ ... _GN having a constant width are formed on the _first pattern layer, and patterns H ₁ ... H _N having a constant width are formed on the second pattern layer. A1, b1, a on the coordinate axis where 0 is an arbitrary point
2, b2 ... aN, bN (normally N is an odd number) are edge positions of the patterns G ₁ ... G _N formed on the first pattern layer, and c1, d1, c2, d2 ... cN, dN (N is an odd number). )
Is the edge position of the patterns H ₁ ... H _N formed on the second pattern layer. Pitch of the pattern _G 1 ... _{G N} (distance between the centers of the pattern) is set to P1, the pattern _{H 1}
The pitch of H _N (the distance between the centers of the patterns) is set to P2, and both pitches P1 and P2 are slightly different.

When the overlay accuracy is 0, the pattern at the center of the pattern row of the first pattern layer, that is, (N + 1) /
The center of the second pattern G (N + 1) / 2 and the center pattern of the pattern row of the second pattern layer, that is, (N +
1) The center of the second pattern H (N + 1) / 2 matches. When the overlay precision is R, R is expressed by the following equation. Next, a method of measuring the overlay accuracy by using the pattern of FIG. 1 by the measuring machine of FIG. 2 will be described.

The parallel plate glass 15 of FIG. 2 is rotated by the control circuit 9 so that the spot light can be scanned on the straight line connecting L1 and L2 of FIG.

The reciprocating mirror 14 is controlled by controlling the interferometer 10.
Is moved to scan the spot light from L1 to L2 in FIG. At the same time, the reflected light data from M1 (corresponding to point 0) to M2 in FIG. 1 is converted into a digital signal by the A / D converter 11 in synchronization with the interferometer signal and is stored in the memory 12. The interval between the data at the address in the memory 12 and the data at the next address corresponds to the interval of the interferometer pulse.

At this time, an automatic gain control (AGC) is applied by an amplifier circuit (not shown) so that the reflected light intensity signal is at an optimum level, but the optimum AGC is not applied by one scanning of the spot light. Repeats the spot scan. Reciprocating mirror 1 for spot scanning
Since 4 is controlled by the interferometer 10, even when spot scanning is performed again, the reflected light intensity data from M1 to M2 in FIG.

In this way, the CPU 13 determines from the reflected light intensity data stored in the memory 12 that the pattern G shown in FIG.
₁ ... each edge position of _{G N a1, b1, a2,} b2 ... a
Calculate N and bN. Specifically, by utilizing the fact that the reflected light intensity changes in the vicinity of the edge, the position at which a certain threshold level is reached is detected as the edge from the reflected light intensity in the portion where the patterns G ₁ ... G _N are not present, and the edge position is determined. Ask.

Next, the parallel plate glass 15 is rotated by the control circuit 9 so that the spot light can scan on the straight line connecting L3 and L4 in FIG. The interferometer 10 is controlled and the reciprocating mirror 14 is moved to scan the spot light from L3 to L4 in FIG. At the same time, the reflected light data from M3 in FIG. 1 (corresponding to the 0 point like M1 as is clear in the figure) to M4 is synchronized with the interferometer signal in the A / D converter 11
Is converted into a digital signal and is taken into the memory 12. The interval between data at one address and the next address in the memory 12 corresponds to the interval of interferometer pulses.

At this time, an automatic gain control (AGC) is applied by an amplifier circuit (not shown), but if the optimum AGC is not applied in one scanning of the spot light, the spot scanning is repeated. Since the reciprocating mirror 14 that performs spot scanning is controlled by the interferometer 10, even when performing spot scanning again, M3 to M in FIG.
The reflected light intensity data up to 4 is stored in the memory 12.

In this way, the CPU 13 causes the pattern H ₁ ... H based on the reflected light intensity data stored in the memory 12.
_N edge positions c1, d1, c2, d2 ... cN, dN
To calculate.

Finally, the CPU 13 calculates the overlay accuracy R and displays the measured value on a CRT or the like (not shown).

The calculated patterns G ₁ ... G _N , H ₁ ... H
Edge positions a1, b1, c1, d1, a2, b of _N
2, c2, d2 ... aN, bN, cN, dN all represent the distance from the 0 point as is clear from the above description, and therefore the overlay accuracy R is expressed by the following equation. The measuring machine may calculate any one of R1, R2 ... RN and use it as the measured value. R = R1 due to subtle changes in edge roughness, optical noise, and grain effects.
= R2 = R3 = ... = RN may not be obtained in some cases.
The measured value may be R = (R1 + R2 + R3 + ... + RN) / N.

FIG. 3 is a block diagram of another measuring machine for carrying out the overlay accuracy measuring method according to one embodiment of the present invention. The same parts as those of the embodiment of FIG. 2 are designated by the same reference numerals and the description thereof will be omitted.

The stage 6 has a moving function.
A control circuit 9 for supplying a signal for controlling the moving direction and the moving amount of the stage 6 to the stage 6 is connected to the, and an interferometer 10 is connected.

The light flux of the laser emitted from the light source 1 is expanded by a beam expander (not shown), and then the condenser lens 2 is used.
Then, the light passes through the beam splitter 3, is reflected by the total reflection mirror 4, enters the objective lens 5, is condensed by the objective lens 5, and forms an image on the sample 7. Stage 6 is interferometer 10
The position control is performed by the control circuit 9, and the spot irradiation position on the sample 7 is moved by moving the stage 6. The signal photoelectrically converted by the reflected light detector 8 is converted into a digital signal by the A / D converter 11,
The signals are sequentially captured in the memory 12 in synchronization with the interferometer signal of the interferometer 10. The interferometer 10 also controls the stage 6, and the address of the memory 12 where the data is captured and the sample 7
The image forming position of is determined uniquely. The data taken into the memory 12 is processed by the CPU 13 to determine the edge position.

Next, a method of measuring overlay accuracy by using the pattern of FIG. 1 by the measuring machine of FIG. 3 will be described.

The stage 6 is moved to L1 in FIG. The interferometer 10 is controlled, the stage 6 is moved, and the spot light is scanned by moving the stage 6 from L1 to L2 in FIG. At the same time, from M1 (corresponding to 0 point) in FIG.
The reflected light data up to 2 is converted into a digital signal by the A / D converter 11 in synchronization with the interferometer signal, and the memory 12
Take in. The interval between the data at the address in the memory 12 and the data at the next address corresponds to the interval of the interferometer pulse.

At this time, an automatic gain control (AGC) is applied by an amplifier circuit (not shown), but if optimum AGC is not applied in one scan by moving the stage 6, spot scanning is repeated. Since the stage 6 is controlled by the interferometer 10, even when spot scanning is performed by moving the stage again, the reflected light intensity data from M1 to M2 in FIG.

In this way, the CPU 13 causes the patterns G ₁ ... G to be obtained from the reflected light intensity data stored in the memory 12.
_N positions a1, b1, a2, b2 ... AN, bN are calculated.

Next, the stage 6 is moved to L3 in FIG.
The spot light is scanned by controlling the interferometer 10 and moving the stage 6 from L3 to L4 in FIG. at the same time,
The reflected light data from M3 (corresponding to 0 point like M1 as is clear in the figure) to M4 in FIG. 1 is converted into a digital signal by the A / D converter 11 in synchronization with the interferometer signal, and stored in the memory 12. Take in. The interval between the data at the address in the memory 12 and the data at the next address corresponds to the interval of the interferometer pulse.

At this time, an automatic gain control (AGC) is applied by an amplifier circuit (not shown), but if optimum AGC is not applied in one spot scan by stage movement, the spot scan is repeated. Since the stage 6 is controlled by the interferometer 10, even when spot scanning is performed again, the reflected light intensity data from M3 to M4 in FIG.

In this way, the CPU 13 causes the pattern H ₁ ... H based on the reflected light intensity data stored in the memory 12.
_N edge positions c1, d1, c2, d2 ... cN, dN
To calculate.

Finally, the CPU 13 calculates the overlay accuracy R and displays the measured value on a CRT or the like (not shown).

The calculated patterns G ₁ ... G _N , H ₁ ... H
Edge positions a1, b1, c1, d1, a2, b of _N
2, c2, d2 ... aN, bN, cN, dN all represent the distance from the 0 point, as is clear from the above description.
The overlay accuracy R is represented by the same equation as the above equation.

[0043]

As described above, according to the overlay accuracy measuring method of the present invention, the overlay accuracy can be measured by a measuring machine using a visual measurement pattern. It is easy to calibrate the measured value and the measured value visually, and the pattern dedicated to the measuring machine is not required.

[Brief description of drawings]

FIG. 1 is a diagram showing a pattern for visually measuring overlay accuracy.

FIG. 2 is a block diagram of a measuring machine for carrying out the overlay accuracy measuring method according to an embodiment of the present invention.

FIG. 2 is a block diagram of another measuring machine for carrying out the overlay accuracy measuring method according to an embodiment of the present invention.

FIG. 4 is a diagram showing a pattern for visually measuring overlay accuracy.

FIG. 5 is a diagram showing a pattern dedicated to a measuring machine.

[Explanation of symbols]

A ₁ ... A _N pattern B ₁ ... B _N pattern a1 ... aN pattern edge position b1 ... bN pattern edge position

Claims (1)

[Claims] 1. A first pattern group in which a first pattern having a width equal to that of a first pattern layer is formed at a first pitch, and a second pattern layer having a width equal to that of a second pattern layer adjacent to the first pattern layer. A second pattern group formed by forming a second pattern at a second pitch slightly different from the first pitch and partially overlapping with the first pattern in a direction orthogonal to the arrangement direction; In the method of measuring the overlay accuracy of the first pattern layer and the second pattern layer using a vernier-type overlay accuracy measurement pattern, the scan pattern is formed by scanning only the first pattern group with a light beam. Then, a first step of receiving the reflected light to determine the edge position of the first pattern, and moving the light beam in a direction orthogonal to the arrangement direction of the first pattern, and then the second pattern group A second step of scanning only the light beam with the light beam and receiving the reflected light to determine the edge position of the second pattern; and the first pattern obtained in the first step and the second step. A method of measuring overlay accuracy, which comprises determining an overlay accuracy of the first pattern layer and the second pattern layer based on an edge position of the second pattern.