JPH0894315A - Alignment method, light projection exposure device by the method, and position deviation measuring instrument - Google Patents

Abstract

PURPOSE: To improve alignment accuracy without adding any hardware by calculating correlation results using a template corresponding to an alignment mark and a template where either a right half or a left half is shifted by one point. CONSTITUTION: Light from an alignment lighting device 11 is applied to an alignment mark on a wafer 5 mounted on an XY stage 7 and the reflection light is subjected to photoelectric conversion by an alignment sensor 12, is A/D-converted by an A/D converter 101, and then is stored at a storage 102. A first operation block 103 performs correlation operation with the template for only the edge part of a mark to calculate a first correlation result. A second operation block 104 shifts the right half or left half within a range where correlation operation is made by one point and performs correlation operation with the template to calculate a second correlation result. A mark center detection block 106 synthesizes the first and second correlation results to create a third correlation result and the mark center is calculated by using it.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a position measuring method performed in a semiconductor manufacturing apparatus, and more particularly, to an automatic alignment performed as a wafer position adjustment in a step-and-repeat type exposure apparatus (hereinafter referred to as a stepper). The present invention relates to a position measuring method used in.

[0002]

2. Description of the Related Art FIG. 9 is a block diagram showing a structure of an exposure apparatus which has been conventionally used. A conventional stepper position measuring method will be described below with reference to FIG.

The wafer 1005 mounted on the XY stage 1007 via the wafer chuck 1006 is illuminated with the illumination light from the exposure illumination device 1001 which is formed into an exposure pattern by passing through the shutter 1002, the reticle 1003 and the projection exposure lens 1004. Is irradiated.

An alignment mark (not shown) on the wafer 1005 mounted on the XY stage 1007 is illuminated by an alignment mark illuminating device 1011 which is a light source. This illumination light is reflected, and the reflected light is reflected by the alignment sensor 1
The light is received at 012 and photoelectric conversion is performed. An area sensor such as a CCD camera, a CCD line sensor, a photodiode, or the like is used as the alignment sensor 1012.

A signal photoelectrically converted by the alignment sensor 1012 is A / D converted by an A / D conversion device 1101 and stored in a storage device 1102 in the processing device as a plurality of pieces of digital pixel information. The stored pixel information is subjected to signal processing in the calculation block 1103, the center of the alignment mark is obtained in the peak position detection block 1105 from the signal processing result, the wafer position is calculated, and the wafer position is sent to the stepper controller 1014. . When the wafer position is calculated, the stepper control device 1014 monitors the stage position with the interferometer 1009,
The shutter 1002 was opened when the position of the wafer 1005 and the reticle 1003 were accurately aligned by driving 08, and the circuit pattern of the reticle 1003 was exposed on the wafer 1005.

The following method has been adopted as the above-mentioned signal processing method.

(1) The alignment mark taken in from the CCD camera is projected in the long axis direction, and template matching is performed by the obtained signal.

(2) Template matching is performed by the signal of the alignment mark taken in by the line sensor.

Template matching is a correlation calculation between a signal collected by a sensor and a template held in advance in a processing device, and the position with the highest correlation is detected as the center of the alignment mark. Template matching is expressed by the following equation.

[0010]

[Equation 1] Here, S is a signal collected by the sensor, T is a template, and E is a correlation result. The relationship among the signal S, the template T, and the correlation value E is shown in FIG. 10 (A).

As a template, the left half of the signal sampled by the sensor can be folded back for correlation calculation.

[0012]

[Equation 2] The formula (2) has an advantage that the template size and the template information need not be specified. This is because the mark has a bilaterally symmetrical shape.

The center of the mark is the point where E (x) becomes maximum. Further, in order to perform the calculation at higher speed, the formula (2) is transformed into the formula (3).

[0014]

[Equation 3] In the case of the equation (3), the value becomes the minimum at the position having the correlation. Therefore, it suffices to search for the position where the value is the minimum at the symmetrical mark center. The above calculation is based on the calculation block 1103 described above.
In the method of equation (1), the template T is stored in the template storage block 1104.

[0015]

As described above, in the conventional alignment method, template matching processing is performed on the alignment mark captured by the sensor.

However, the signal captured by the sensor changes when the registration mark is coated with a resist or the like, and the signal width corresponding to the line width becomes thicker.
It becomes thin. For this reason, it may not match the template already held, and the position of the highest correlation and the true center position may deviate by 1/2 pixel. If the exposure is performed in a state where the measurement accuracy is deteriorated, the exposure accuracy is deteriorated, and as a result, the production yield of the semiconductor is lowered.

The cause of the above-mentioned 1/2 pixel shift will be described with reference to FIG. FIG. 10 (A) shown above is
The case where the size of the waveform S is 2k + 1 points which is the same as that of the template T is shown. In this case, the correlation E has only one peak. However, k is a natural number, and in the case of FIG. 3, k = 11.

FIG. 10B shows a case where the size of the waveform S'is 2k + 2 points, which is one point larger than the template T. In this case, the original peak position is the midpoint between the two peak positions, but the peaks of the correlation degree E ′ appear at the positions adjacent to the two peaks. This is because there are two
This is because even a small point among the points is determined as the center of the mark, and as a result, it is displaced by 1/2 pixel.

As a method of reducing the amount of 1/2 pixel shift, there is a method of increasing the pixel resolution. The pixel resolution can be improved by doubling the sampling frequency for A / D conversion and doubling the amount of memory. However, in this case, there is a problem that the scale of the hardware becomes large and the manufacturing cost becomes high.

The present invention has been made in view of the problems of the above-mentioned conventional technique, and realizes a positioning method capable of improving the measurement accuracy without increasing the manufacturing cost. The purpose is to do.

[0021]

According to the alignment method of the present invention, a position for confirming the position of a predetermined object is detected by two-dimensionally detecting an alignment mark provided on a predetermined object and performing template matching. A matching method, wherein a first correlation calculation is performed using a template corresponding to the alignment mark to calculate a first correlation result, and a right half or a left half of a range in which the correlation calculation of the digital signal is performed is performed. 1
Point-shifted to perform a second correlation operation on the template to calculate a second correlation result, and alternately synthesize the first correlation result and the second correlation result to create a third correlation result. , The center of the alignment mark is calculated from the third correlation result.

According to a second alignment method of the present invention, a template corresponding to the alignment mark is used to perform a first correlation operation to calculate a first correlation result, and a right half or a left half of the template is calculated. A second template shifted by 1 point is created, a second correlation result is calculated by performing a second correlation calculation with the digital signal of the alignment mark, and the first correlation result and the second correlation result are calculated. Are alternately combined to create a third correlation result, and the center of the alignment mark is calculated from the third correlation result.

According to the third alignment method of the present invention, half of the two-dimensionally detected alignment marks is regarded as a template, and a first folded correlation calculation is performed with the remaining half of the digital signals to perform the first alignment. The second correlation result is calculated by calculating the correlation result, shifting the range in which the correlation calculation of the template and the remaining half of the alignment mark detected two-dimensionally is performed by one point, and performing the second folded correlation calculation. The third correlation result is calculated by alternately synthesizing the first correlation result and the second correlation result, and the center of the alignment mark is calculated from the third correlation result. .

The fourth alignment method of the present invention regards half of the two-dimensionally detected alignment marks as a template, and integrates the difference in symmetrical position with the digital signal of the other half. Calculation is performed to calculate a first calculation result, and the range in which the template and the remaining half of the two-dimensionally detected alignment mark are calculated is 1
Second point-shifting and integrating the difference between symmetrical positions
Is performed to calculate a second operation result, the first operation result and the second operation result are alternately combined to create a third operation result, and the position adjustment is performed from the third operation result. The feature is that the center of the mark is calculated.

The image pickup means for picking up the alignment mark on the wafer surface and the alignment mark on the reticle surface according to the present invention, and the image pickup means for projecting and exposing the pattern on the reticle surface onto the resist on the wafer surface. A projection exposure apparatus equipped with an alignment mechanism that relatively positions the reticle and the wafer using the obtained information, an image pickup means for picking up an alignment mark on the wafer surface, and a reticle surface on a resist on the wafer surface. A projection exposure apparatus provided with an alignment mechanism for performing relative alignment between the reticle and the wafer by using information obtained by the imaging means to project and expose the above pattern, and a positional deviation measurement mark on the wafer surface. An image pickup means for picking up an image is provided, and the position of the position shift measurement mark is detected using the information obtained by the image pickup means to measure the shift amount. Also characterized by performing alignment using any of the methods described above in any of that positional deviation measuring device.

[0026]

In the alignment method of the present invention configured as described above, in order to eliminate the shift amount of 1/2 pixel without adding hardware, specifically, the following signal processing is performed. Be seen.

[0027]

[Equation 4] F is created by synthesizing the calculation results E1 and E2. The synthesis of F is shown in equation (6).

[0028]

[Equation 5] Equation (4) is a conventional correlation calculation. Expression (5) is a correlation calculation in which the position where the right half of the template is applied is shifted by one pixel. By synthesizing the equations (4) and (5), the correlation calculation with 1/2 pixel precision is realized.

[0029]

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

First Embodiment FIG. 1 is a block diagram showing the arrangement of a first embodiment of the exposure apparatus according to the present invention.

With reference to FIG. 1, description will be given of a method of taking in the alignment mark, a signal processing device and a processing method in this embodiment.

In this embodiment, the wafer 5 mounted on the XY stage 7 via the wafer chuck 6 passes through the shutter 2, the reticle 3 and the projection exposure lens 4 to form an exposure illumination device 1 having an exposure pattern. The illumination light from is emitted.

The alignment mark (not shown) is provided on the wafer 5 mounted on the XY stage 7.
The light emitted from the alignment mark illuminating device 11, which is a light source, passes through the half mirror 10, the mirror 13, and the reduction type projection exposure lens 4, and illuminates the alignment mark. The image of the alignment mark generated by the illumination is reflected by the wafer 5, passes through the half mirror 10 along the same path as the incidence, and is incident on the alignment sensor 12. The alignment sensor 12 is composed of an area sensor such as a CCD camera, a CCD line sensor, a photodiode or the like, and the image of the alignment mark received by these is photoelectrically converted.

The alignment mark may be, for example, a combination of two substances having different reflectances as shown in FIG. is there. After photoelectric conversion, a signal as shown in the right part of FIG. 2 is output.

The photoelectrically converted signal is processed by the alignment mark detection device 100. The alignment mark detection device 100 includes an A / D conversion device 101 and a storage device 1.
02, first operation block 103, second operation block 10
4, the template holding device 105 and the mark center detection block 106, the processing signal of which is output to the exposure device control device 14, and the exposure device control device 1
Reference numeral 4 controls the operation of the exposure apparatus according to the calculation result.

The signal photoelectrically converted by the alignment sensor 12 is converted into a digital signal by the A / D conversion device 101 and stored in the memory 102 in the alignment mark detection device 100 as a plurality of pieces of digital pixel information.

Enlarging one of the photoelectrically converted information,
The waveform S shown in FIG. 3 is obtained. Each dot in FIG. 3 represents the sampled signal strength. However, when an area sensor such as a CCD camera is used as the alignment sensor 12 and the detection signal by this is photoelectrically converted, it is a signal projected in the longitudinal direction of the mark.

The pixel information stored is subjected to signal processing in the first calculation block 103 and the second calculation block 104, respectively. The mark center detection block 106 obtains the center of the alignment mark based on the calculation result of each calculation block and calculates the position of the wafer.

A method of calculating the wafer position in the mark center detection block 106 will be described with reference to FIG.

The first calculation block 103 performs the following calculation to calculate E1.

[0041]

[Equation 6] The templates are bilateral, so

[0042]

[Equation 7] Expression (7) is a correlation calculation between the signal waveform and the template. The template need only be the mark edge portion. There are two reasons for this. The first reason is that the portions other than the edges do not have mark information, and the second reason is to suppress the influence of the noise component in the area having no information on the correlation calculation. The templates are bilateral, so they are summarized in equation (9). S, T, and E1 are shown in the left part of FIG.

The second calculation block 104 performs the following calculation to calculate E2.

[0044]

[Equation 8] S, T, and E2 are shown in the right part of FIG.

In the mark center detection block 106, the calculation results E1 and E2 of each calculation block are combined to create F. The synthesis of F is shown in equation (11).

[0046]

[Equation 9] The peak of F (x ₂ ) is the mark center position. E1 and E
The state where 2 is synthesized is shown in the lower part of the figure.

Comparing E1 and E2, the first term is the same. In the second term, the right template is hung outside by one pixel. When x is arranged in order of x-1, x, x + 1 for E1 and E2,

[0048]

[Equation 10] Referring to the above equations, it can be seen that E2 interpolates between E1. In the description using FIG. 3, the template is applied to the outside by one pixel in the correlation calculation of E2, but a similar result can be obtained by applying it to the inside by one pixel. Of course, there are some variations of the calculation method, but the idea is the same.

When the position of the wafer is calculated by the mark center detection block 106, the exposure apparatus controller 14
Drives the stage driving device 8 while monitoring the position of the XY stage 7 by the laser interference system 9, and when the position of the wafer 5 and the reticle 3 is accurately aligned, the shutter 2
To expose the circuit pattern of the reticle 3 on the wafer 5.

The above-mentioned first operation block 103,
The second calculation block 104 and the mark center position detection block 106 can be realized by one general-purpose processor.

Second Embodiment As a second embodiment of the present invention, the first operation block 10
3, the operation content of the second operation block 104 A method of performing the correlation operation method with the template side shifted will be described. Since the process for photoelectrically converting the alignment mark, the stage control, and the exposure method are the same as those in the first embodiment shown in FIG. 1, the description thereof will be omitted.

The first calculation block 103 performs the following calculation to calculate E1.

[0053]

[Equation 11] Expression (18) is the same as Expression (9) in the first embodiment.

The second calculation block 104 performs the following calculation to calculate E2.

[0055]

[Equation 12] In the mark center detection block, F is created by synthesizing the calculation results E1 and E2 of each calculation block. The synthesis of F is shown in equation (20).

[0056]

[Equation 13] The peak of F (x ₂ ) is the mark center position. Comparing the operation results E1 and E2 of each operation block, the first term is the same. The second term is multiplied by shifting the right template inward by one pixel. The reason why the right template is shifted inward by one pixel in the second embodiment is that it corresponds to the first embodiment. As described in the first embodiment, a similar result can be obtained even if the right template is shifted outward by one pixel, and there are several variations of this equation, but the idea is the same.

Also in the present embodiment, the first operation block 103 and the second operation block 1 are the same as in the first embodiment.
04 and the mark center position detection block 106 can be realized by one general-purpose processor.

Third Embodiment As a third embodiment of the present invention, the first operation block 10
3. A method of performing the calculation contents of the second calculation block 104 by the return correlation calculation method will be described. The process up to photoelectric conversion of the alignment mark, the stage control, and the exposure method in this embodiment are the same as those in the first embodiment, and the description thereof will be omitted.

When one of the photoelectrically converted signal waveforms is enlarged, the waveform S shown in FIG. 2 is obtained. The signal strength at which each dot is sampled is shown. The left half of the signal waveform is the folding template. The template need only be the mark edge portion. There are two reasons for suppressing the noise component in the area having no information, which does not have the mark information in the portion other than the edge, from affecting the correlation calculation. The correlation calculation is shown in the following equation.

The first calculation block 103 performs the following calculation to calculate E1.

[0061]

[Equation 14] Expression (21) is a correlation calculation in which the left half of the signal waveform (waveform S shown in FIG. 3) is regarded as the template. The template need only be the mark edge portion. There are two reasons. Firstly, the portions other than the edges do not have mark information, and secondly, the noise components in the area having no information affect the correlation calculation.

The second calculation block 104 performs the following calculation to calculate E2.

[0063]

(Equation 15) In the mark center detection block 106, the calculation results E1 and E2 of each calculation block are combined to create F. The synthesis of F is shown in equation (23).

[0064]

[Equation 16] The peak of F (x ₂ ) is the mark center position.

Also in this embodiment, as in the first embodiment, the first operation block 103 and the second operation block 1
04 and the mark center position detection block 106 can be realized by one general-purpose processor.

Fourth Embodiment As a fourth embodiment of the present invention, the first operation block 10
3. A method of implementing the operation contents of the second operation block 104 as a high-speed correlation operation method for performing high-speed correlation operation in the second embodiment will be described. The process up to photoelectric conversion of the alignment mark, the stage control, and the exposure method in this embodiment are the same as those in the first embodiment, and the description thereof will be omitted.

When one of the photoelectrically converted signal waveforms is enlarged, a waveform S shown in the figure is obtained. The signal strength at which each dot is sampled is shown. The left half of the signal waveform is the folding template. The template need only be the mark edge portion. There are two reasons for suppressing the noise component in the area having no information, which does not have the mark information in the portion other than the edge, from affecting the correlation calculation. The correlation calculation is shown in the following equation.

[0068]

[Equation 17] Expressions (24) and (25) integrate the difference between the signals at the folding-symmetrical positions. In the third embodiment, multiplication was used,
In a microprocessor or the like, the subtraction processing time is shorter. In the mark center detection block 106, the calculation results E1 and E2 of each calculation block are combined to create F. F
The synthesis of is shown in formula (26).

[0069]

(Equation 18) However, in the mark center detection block 106, F
The minimum value of (x ₂ ) is searched for, and that position is the mark center position. In addition, F (x ₂ ) is the reciprocal F ′ (x ₂ )

[0070]

[Formula 19] Then, the center of the mark can be obtained from the peak.

Also in this embodiment, as in the first embodiment, the first operation block 103 and the second operation block 1
04 and the mark center position detection block 106 can be realized by one general-purpose processor.

Fifth Embodiment FIG. 4 is a block diagram showing the arrangement of the fifth embodiment of the exposure apparatus according to the present invention.

With reference to FIG. 4, description will be given of a method of taking in the alignment mark, a signal processing device and a processing method in this embodiment.

In this embodiment, the alignment mark on the wafer passed through the reticle is sampled.

This embodiment is different from the first embodiment shown in FIG. 1 in that the reticle 3 receives an instruction from the exposure apparatus controller 14.
The reticle driving device 301 for driving the lens and the lens magnification driving device 305 for setting the magnification of the projection exposure lens 4 are provided, and the half mirror 10 and the mirror 1 in the first embodiment are provided.
3, the alignment mark illuminating device 11 and the alignment sensor 12 are respectively provided with a half mirror 302 and a mirror 3.
00, an alignment mark illumination device 304, and an alignment sensor 303. Since the other structure is the same as that of the first embodiment shown in FIG. 1, the same reference numerals as those in FIG.

The alignment mark of the reticle 3 in this embodiment exists as a chrome pattern on the reticle 3. An alignment mark for the wafer is formed on the wafer 5 mounted on the XY stage 7.

Positioning mark illuminating device 30 which is a light source
The light emitted from 4 is the half mirror 302 and the mirror 30.
0, the reticle 3, and the reduction projection lens 4 to illuminate the alignment mark on the wafer 5. The image of the alignment mark is reflected by the wafer 5, follows the same path again, passes through the reticle 3 and the half mirror 302, and enters the alignment sensor 303. The image of this alignment mark is
Area sensor such as CCD camera or alignment sensor 30 such as CCD line sensor or photodiode
Photoelectric conversion is performed by 3.

The alignment mark of the reticle 3 is illuminated by the reflected light from the wafer 5. At the time of the illumination, the chrome pattern forming the alignment mark is shielded from light, and a dark image is formed on the alignment sensor 303. FIG. 5 shows an example of an image of the alignment mark taken by the CCD camera. FIG. 5 also shows a composite image of the left optical system (not shown).

The photoelectrically converted signal is sent to the A / D converter 1
A / D conversion is performed at 01 and stored in the storage device 102 in the processing device as a plurality of pixel information.

When one of the photoelectrically converted information is enlarged, the waveform S shown in the figure is obtained. The signal strength at which each dot is sampled is shown. However, when photoelectrically converted by an area sensor such as a CCD camera, it is a signal projected in the longitudinal direction of the mark.

In the storage device 102, X and Y on the right and left sides
Marks (XL, YL, XR, YR) and a total of four signals are stored.

The four pieces of stored pixel information are independently stored in the first
The calculation block 103 and the second calculation block 104 perform signal processing, and the mark center detection block 106 obtains the center of the alignment mark from the calculation results of both. The calculation at this time is the center (XLW) of the alignment mark on the wafer 5.
C), the first reticle mark center (XLR1) and the second reticle mark center (XLR2) are detected. In this way, the wafer 5 and the reticle 3 are independently measured, and the relative positional deviation amount is measured.

As the method of calculating the mark center position, any of the methods described in Embodiments 1 to 4 can be used.

When the relative deviation amounts of the XL, YL, XR, and YR marks are measured, the exposure apparatus controller 14
The shift component, the rotation component, and the magnification component in the X and Y directions are calculated, and while the position of the XY stage 7 is monitored by the interference system 9, the reticle position driving device 301 or the stage driving device 8 is driven to shift and rotate components. The correction is performed, and the wafer rotation device 6 is driven to correct the rotation component. The magnification component is corrected by driving the lens magnification driving device 305. When the positions of the reticle 3 and the wafer 5 are accurately aligned, the shutter 2 is opened and the circuit pattern of the reticle 3 is exposed on the wafer 3.

Sixth Embodiment FIG. 6 is a block diagram showing the arrangement of a sixth embodiment of the exposure apparatus according to the present invention.

With reference to FIG. 6, description will be given of a method of taking in the alignment mark, a signal processing device and a processing method in this embodiment.

In this embodiment, the reduction projection lens is not passed,
The alignment mark on the wafer is sampled using an external microscope.

This embodiment is different from the first embodiment shown in FIG. 1 in that an off-axis microscope 401 is provided, and the half mirror 10, the mirror 13, the alignment mark illuminating device 11 and the alignment sensor in the first embodiment are provided. Each of 12 is a half mirror 403, a mirror 402, an alignment mark illuminating device 405, and an alignment sensor 404.
It is what Since the other structure is the same as that of the first embodiment shown in FIG. 1, the same reference numerals as those in FIG.

Wafer 5 mounted on XY stage 7
An alignment mark for the wafer is formed on the top.

Positioning mark illuminator 40 which is a light source
The light emitted from 5 is the half mirror 403 and the mirror 40.
The alignment mark on the wafer 5 is illuminated through the 2-off axis microscope 401. The image of the alignment mark is reflected by the wafer 5, follows the same path again, and enters the alignment sensor 303. The image of the alignment mark is photoelectrically converted by an area sensor such as a CCD camera or an alignment sensor 404 such as a CCD line sensor or a photodiode.

As the method of calculating the mark center position, any of the methods described in Embodiments 1 to 4 can be used.

Seventh Embodiment FIG. 7 is a block diagram showing the arrangement of a sixth embodiment of the exposure apparatus according to the present invention.

With reference to FIG. 7, the method of taking in the alignment mark, the signal processing apparatus and the processing method in this embodiment will be described.

In this embodiment, the method according to the present invention is applied to a deviation amount measuring device. The deviation amount measuring device aims to measure the deviation amount measuring mark formed on the wafer surface after the exposure process is performed while the pattern on the reticle surface is aligned by the exposure device and the resist and the like are developed. The wafer 501 mounted on the wafer chuck 502 on the XY stage 503 is irradiated with the illumination light from the deviation amount measurement mark illumination device 505 via the half mirror 507 and the microscope 508. A displacement amount measurement mark as shown in FIG. 8 is formed on the upper surface of the wafer 501, and the image of the displacement amount measurement mark is reflected by the wafer 501, traces the same path again, passes through the half mirror 507, and then the CCD.
Area sensor such as a camera or two sets of X and Y CCDs
It is incident on the deviation amount observing and imaging device 504 such as a line sensor and a photodiode, and photoelectrically converted.

Since the structure of the deviation amount detecting device 500 is the same as that of the first embodiment shown in FIG. 1, its explanation is omitted. The detection result of the alignment mark detection device 100 is output to the displacement amount measurement device control device 509, and the displacement amount measurement device control device 509 controls the XY stage 503 via the stage drive device 506 based on the detection result. Also. Output to the output device 510.

The deviation amount detection performed in this embodiment will be described.

In the image of the shift amount measurement mark shown in FIG. 8, the center of the outer mark (OUTM) and the center of the inner mark (INM) are signaled by any one of the first to fourth embodiments. Waveform XMEA, YMEA X axis, Y
Each axis is obtained, and these differences are sent to the deviation amount measuring device control device 509 to be stored therein. The shift amount measuring device control device 509 controls the X-axis through the stage driving device 506.
The Y stage 503 is driven and other misalignment amount measurement marks are measured. Finally, statistics and the like are calculated and transferred to the output device 510 together with a plurality of stored measurement values,
Output.

[0098]

Since the present invention is configured as described above, it has the following effects.

In any of the methods described in claims 1 to 4, it is possible to detect with double the precision as compared with the conventional method. Up to now, the detection resolution is 1/2
There was a place to rely on hardware to double the sampling frequency in order to achieve this, but using this method, the precision can be doubled while inheriting the conventional hardware assets. Moreover, in the conventional processing method, if the operation block is executed by software, the accuracy can be improved to 2 by simply replacing the conventional software.
Can be doubled. It is an extremely inexpensive method for improving performance. Further, since the alignment accuracy is insufficient, it can be easily introduced into a semiconductor manufacturing process where high-precision processing cannot be performed, and the processing accuracy can be doubled while using the conventional exposure apparatus.

The effect peculiar to the present invention is not limited to the semiconductor exposure apparatus according to any one of claims 5 to 12,
It can also be used for mark position detection of a positional deviation inspection device after exposure as described in claim 16. Also in this case, the measurement accuracy can be easily doubled.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a projection exposure apparatus in which a first embodiment to a fourth embodiment of the present invention are carried out.

2A and 2B are diagrams showing alignment marks photoelectrically converted by the projection exposure apparatus shown in FIG. 1 and a signal waveform.

FIG. 3 is a diagram for explaining a calculation method of the present invention.

FIG. 4 is a diagram showing a configuration of a projection exposure apparatus according to a fifth embodiment of the present invention.

FIG. 5 is a diagram showing alignment marks imaged by a projection exposure apparatus according to a fifth embodiment of the present invention.

FIG. 6 is a diagram showing a configuration of a projection exposure apparatus according to a sixth embodiment of the present invention.

FIG. 7 is a diagram showing a configuration of a projection exposure apparatus according to a seventh embodiment of the present invention.

FIG. 8 is a diagram showing marks imaged by a deviation amount measuring device according to a seventh embodiment of the present invention.

FIG. 9 is a diagram for explaining a conventional example of a projection exposure apparatus.

FIG. 10 is a diagram showing a problem of a conventional correlation calculation.

[Explanation of symbols]

1 Exposure Illuminator 2 Shutter 3 Reticle 4 Projection Exposure Lens 5,501 Wafer 6,502 Wafer Chuck 7,503 XY Stage 8,506 Stage Driving Device 9 Laser Interferometer 10,302,403,507 Half Mirror 11,304,405 , 505 Alignment mark illumination device 12, 303, 404 Alignment sensor 13, 300, 402 Mirror 14 Exposure device control device 100, 200 Alignment mark detection device 101, 201 A / D conversion device 102, 202 Storage device 103 First operation Block 104 Second operation block 105 Template holding device 106, 206 Mark center detection block 301 Reticle drive device 401 Off-axis microscope 500 Deviation amount detection device 504 Deviation amount observation / imaging device 508 Microscope 509 Deviation amount measuring device control device 510 output device T template S signal of alignment mark E correlation calculation result F correlation calculation result after combination

Claims (16)

[Claims] 1. A positioning method for two-dimensionally detecting an alignment mark provided on a predetermined object to perform template matching to confirm the position of the predetermined object, which corresponds to the alignment mark. The first correlation calculation is performed using the template and the first correlation result is calculated, and the right half or left half of the range in which the correlation calculation of the digital signal is performed is shifted by 1 point, and the second correlation is calculated using the template. An operation is performed to calculate a second correlation result, the first correlation result and the second correlation result are alternately combined to create a third correlation result, and an alignment mark is generated from the third correlation result. A method for detecting an alignment mark, which comprises calculating the center of the alignment mark. 2. A positioning method for two-dimensionally detecting an alignment mark provided on a predetermined object to perform template matching to confirm the position of the predetermined object, which corresponds to the alignment mark. The first template
Correlation calculation is performed to calculate the first correlation result, a second template is created by shifting the right half or left half of the template by 1 point, and the second correlation calculation is performed with the digital signal of the alignment mark. Is performed to calculate a second correlation result, the first correlation result and the second correlation result are alternately combined to create a third correlation result, and a registration mark of the alignment mark is generated from the third correlation result. A registration mark detection method characterized by calculating a center. 3. A positioning method for two-dimensionally detecting a positioning mark provided on a predetermined object to perform template matching to confirm the position of the predetermined object. One half of the registered alignment marks is regarded as the template, and the other half is used as the first
Is performed to calculate the first correlation result, and the range for performing the correlation calculation of the remaining half of the template and the two-dimensionally detected alignment mark is shifted by 1 point to obtain the second return correlation calculation. Is performed to calculate a second correlation result, the first correlation result and the second correlation result are alternately combined to create a third correlation result, and a registration mark of the alignment mark is generated from the third correlation result. A registration mark detection method characterized by calculating a center. 4. A positioning method for two-dimensionally detecting a positioning mark provided on a predetermined object to perform template matching to confirm the position of the predetermined object, wherein the two-dimensional detection is performed. One half of the registered alignment marks is regarded as a template, and a first calculation for integrating a difference in symmetrical position with the digital signal of the other half is performed to calculate a first calculation result. The range that is calculated with the remaining half of the alignment mark that has been detected, is shifted by 1 point, the second calculation that integrates the difference between the symmetrical positions is performed, and the second calculation result is calculated. The alignment mark detection method characterized by alternately synthesizing the calculation result and the second calculation result to create a third calculation result, and calculating the center of the alignment mark from the third calculation result. 5. An image pickup means for picking up an alignment mark on a wafer surface and an alignment mark on a reticle surface, and an image pickup means for projecting and exposing a pattern on a reticle surface onto a resist on a wafer surface. A projection exposure apparatus including an alignment mechanism that performs relative alignment between the reticle and the wafer by using information obtained by performing the first correlation calculation using a template corresponding to the alignment mark to obtain a first correlation result. And a first calculation block for calculating the signal, and a right half or left half of the range for performing the correlation calculation of the signal imaged by the image pickup means is shifted by 1 point to perform the second correlation calculation with the template. The second operation block for calculating the second correlation result, and the first correlation result and the second correlation result are alternately combined to create a third correlation result. And a mark center detection block that calculates the center of the alignment mark from the correlation result of 1. 6. An image pickup means for picking up an alignment mark on a wafer surface and an alignment mark on a reticle surface, and an image pickup means for projecting and exposing a pattern on a reticle surface onto a resist on a wafer surface. A projection exposure apparatus including an alignment mechanism that performs relative alignment between the reticle and the wafer by using information obtained by using the template,
A first calculation block for calculating the first correlation result by performing the correlation calculation of 1. and a second template in which the right half or the left half of the template is shifted by 1 point, and a digital signal of the alignment mark is generated. A second operation block for performing a second correlation operation to calculate a second correlation result, and alternately synthesizing the first correlation result and the second correlation result to create a third correlation result, And a mark center detection block for calculating the center of the alignment mark from the third correlation result. 7. An image pickup means for picking up an alignment mark on a wafer surface and an alignment mark on a reticle surface, and an image pickup means for projecting and exposing a pattern on a reticle surface onto a resist on a wafer surface. In a projection exposure apparatus including an alignment mechanism that performs relative alignment between the reticle and the wafer by using the information obtained, half of the alignment marks imaged by the imaging means is regarded as a template, and the remaining half of the digital mark is used. A first calculation block for calculating a first correlation result by performing a first folded correlation calculation with a signal, and a range for performing the correlation calculation of the remaining half of the alignment mark imaged by the template and the image pickup means is 1 A second calculation block for calculating a second correlation result by performing point shift and second folded correlation calculation; and the first correlation result. A projection including a mark center detection block that creates a third correlation result by alternately synthesizing the result and the second correlation result, and calculates the center of the alignment mark from the third correlation result. Exposure equipment. 8. An image pickup means for picking up an alignment mark on a wafer surface and an alignment mark on a reticle surface, and an image pickup means for projecting and exposing a pattern on a reticle surface onto a resist on a wafer surface. In a projection exposure apparatus including an alignment mechanism that performs relative alignment between the reticle and the wafer by using the information obtained, half of the alignment marks imaged by the imaging unit is regarded as a template, and the other half is imaged. A first calculation block that calculates a first calculation result by performing a first calculation that integrates a difference in symmetric position with respect to a signal imaged by the image capturing unit, the template, and the image capturing unit. The second calculation result is calculated by shifting the range for calculation with the other half of the alignment mark by 1 point and performing the second calculation that integrates the difference between the symmetrical positions. The second operation block to be output and the first operation result and the second operation result are alternately combined to create a third operation result, and the center of the alignment mark is calculated from the third operation result. A projection exposure apparatus having a mark center detection block. 9. An image pickup means for picking up an alignment mark on a wafer surface, and the information obtained by the image pickup means for projecting and exposing a pattern on the reticle surface onto a resist on the wafer surface. A projection exposure apparatus including an alignment mechanism that performs relative alignment of the first alignment block and a first computation block that performs a first correlation computation with a template corresponding to the alignment mark to compute a first correlation result; A second half is calculated by shifting the right half or the left half within the range for performing the correlation calculation of the signal captured by the image capturing means by one point, and performing the second correlation calculation using the template. The calculation block and the first correlation result and the second correlation result are alternately combined to create a third correlation result. A projection exposure apparatus having a mark center detection block for calculating a heart. 10. An image pickup means for picking up an alignment mark on a wafer surface, and the information obtained by the image pickup means for projecting and exposing a pattern on the reticle surface onto a resist on the wafer surface. In a projection exposure apparatus including an alignment mechanism for performing relative alignment of the first alignment mark with the template corresponding to the alignment mark.
The first calculation block for calculating the first correlation result by performing the correlation calculation of 1 and the second template in which the right half or the left half of the template is shifted by 1 point are created, and are imaged by the imaging means. A second calculation block that calculates a second correlation result by performing a second correlation calculation with the signal, and alternately synthesizes the first correlation result and the second correlation result to create a third correlation result. And a mark center detection block for calculating the center of the alignment mark from the third correlation result. 11. An image pickup means for picking up an alignment mark on a wafer surface, and the information obtained by the image pickup means for projecting and exposing a pattern on the reticle surface onto a resist on the wafer surface. In a projection exposure apparatus including an alignment mechanism for performing relative alignment of the two, the half of the alignment mark imaged by the imaging unit is regarded as a template, and the first folding correlation calculation with the remaining half of the digital signal is performed. A first calculation block for performing a first correlation result, and a range for performing a correlation calculation of the remaining half of the template and the alignment mark imaged by the imaging means is shifted by 1 point to generate a second folded correlation A second operation block that performs an operation to calculate a second correlation result, and the first correlation result and the second correlation result are alternately combined. And a mark center detection block for calculating the center of the alignment mark from the third correlation result, and a projection exposure apparatus. 12. A reticle and a wafer using image pickup means for picking up an alignment mark on a wafer surface, and information obtained by the image pickup means for projecting and exposing a pattern on the reticle surface onto a resist on the wafer surface. In the projection exposure apparatus including an alignment mechanism for performing relative alignment of the image, the half of the alignment mark imaged by the image pickup means is regarded as a template, and the other half of the alignment mark and the signal imaged by the image pickup means. Calculation of a first calculation block that performs a first calculation that integrates a difference between symmetrical positions to calculate a first calculation result, the template, and the remaining half of the alignment mark imaged by the imaging unit. A second operation block that shifts the range in which is performed by 1 point, performs a second operation that integrates a difference between symmetrical positions, and calculates a second operation result; A mark center detection block for alternately synthesizing the first calculation result and the second calculation result to create a third calculation result, and calculating the center of the alignment mark from the third calculation result. A characteristic projection exposure apparatus. 13. A position deviation measurement device, comprising image pickup means for picking up an image of a position deviation measurement mark on a wafer surface, and detecting a position of the position deviation measurement mark by using information obtained by the image pickup means to measure a deviation amount. The device is a template corresponding to the position shift measurement mark.
And a first calculation block for calculating a first correlation result by performing the correlation calculation of 1), and shifting the right half or the left half of the range for performing the correlation calculation of the signal imaged by the image pickup means by one point. A second operation block that performs a second correlation operation by a template to calculate a second correlation result, and the first correlation result and the second correlation result are alternately combined to create a third correlation result. And a mark center detection block that calculates the center of the alignment mark from the third correlation result. 14. A position deviation measuring method, comprising: an image pickup means for picking up an image of a position deviation measurement mark on a wafer surface, and detecting a position of the position deviation measurement mark by using information obtained by the image pickup means to measure a deviation amount. A first calculation block for calculating a first correlation result by performing a first correlation calculation using a template corresponding to the position shift measurement mark; and shifting the right half or left half of the template by 1 point A second calculation block for calculating a second correlation result by performing a second correlation calculation with the signal imaged by the image pickup means, A mark center detection block for alternately synthesizing the two correlation results to create a third correlation result, and calculating the center of the position shift measurement mark from the third correlation result. Deviation measuring device. 15. A position deviation measuring method, comprising: an image pickup means for picking up an image of a position deviation measurement mark on a wafer surface, and detecting a position of the position deviation measurement mark by using information obtained by the image pickup means to measure a deviation amount. In the apparatus, a half of the displacement measurement marks imaged by the image pickup means is regarded as a template, and a first return correlation calculation is performed with a signal imaged by the other half of the image pickup means to perform a first correlation result. And a second calculation block for calculating the second loopback correlation calculation by shifting the range for performing the correlation calculation for the remaining half of the positional deviation measurement mark imaged by the template and the image pickup unit by one point. The second operation block for calculating the second correlation result and the first correlation result and the second correlation result are alternately combined to create a third correlation result, and the alignment matrix is calculated from the third correlation result. And a mark center detection block that calculates the center of the peak. 16. A position deviation measuring method, comprising: an image pickup means for picking up an image of a position deviation measurement mark on a wafer surface, and detecting a position of the position deviation measurement mark by using information obtained by the image pickup means to measure a deviation amount. In the device, a first calculation is performed, in which half of the position shift measurement marks imaged by the image pickup means is regarded as a template, and a difference in symmetrical position with a signal imaged by the other half image pickup means is integrated. The first calculation block for performing the calculation of the first calculation result, the template, and the range for performing the calculation of the remaining half of the positional deviation measurement mark imaged by the imaging means are shifted by 1 point, and the symmetrical position is obtained. A second operation block for performing a second operation for integrating the difference between the first operation result and the second operation result, and alternately combining the first operation result and the second operation result to obtain a third operation result. Create the said And a mark center detection block that calculates the center of the position shift measurement mark from the calculation result of 3.