JPH04196307A - Alignment method - Google Patents

Abstract

PURPOSE:To correct alignment at high speed in a highly precise manner by a method wherein a plurality of detection patterns on a reticle are arranged at optional intervals, and the detection pattern on a wafer stage are arranged by microscopically shifting them from the interval at which detection patterns are arranged on the reticle. CONSTITUTION:An exposing light is made to irradiate on the detecting pattern 9 on a reticle 2, and the pattern is projected on the detection unit 10 provided on a wafer stage 5 through the intermediary of a reduction lens 3. The above- mentioned projected image is detected by the luminous energy detector 12 located in the detection unit 10, a luminous energy detector changing signal is outputted to a main control system 8, the center position of the pattern projected image is computed, the position of the alignment pattern 6, provided on the detection unit 10, is detected by an alignment detector 7, and the positional data is inputted to a main control system 8. At this point, the amount of deviation between the center of the pattern projected image for detection and the position of the alignment pattern 6 is computed. As a result, highly precise alignment can be conducted.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an alignment method and apparatus for relative positioning of a reticle and a wafer in a projection exposure apparatus.

[Conventional technology]

In general, it is preferable to use light at a non-exposure wavelength for alignment, which involves relative positioning of the reticle and wafer in a projection exposure system, in order to prevent the effects of absorption by the resist coated on the wafer in advance. has been done. However, if the projection optical system is affected by heat, atmospheric pressure, etc. and the relative position of the reticle and wafer, the so-called alignment position, deviates, the non-exposure wavelength light used for alignment will be affected by the aberrations of the projection optical system. etc., the amount of deviation of the exposure light is different from that of the exposure light.

Therefore, it is necessary to correct the amount of deviation in alignment detection.

Conventionally, as a means for correcting the amount of deviation in alignment detection, a system as shown in Japanese Patent Application Laid-open No. 7823/1983 has been used. As shown in FIG. 11, this system uses a detection pattern 9 on a reticle 2 of a projection exposure apparatus.
．． A detection unit 10 is provided on the wafer stage 5 for correction.

The detection unit 10 is composed of a light receiving mask 11 and a light amount detector 12, as shown in FIG. The detection mask 11 is provided with a detection pattern 13 having the same shape as the detection pattern 9 provided on the reticle 2, and an alignment pattern 6 is provided next to it. A light amount detector 9 is provided below the detection pattern 13 of the detection mask 11.

In such a configuration, as shown in FIG. 11, the detection unit 10 is moved below the reduction lens 3 (a position where the detection pattern 9 on the reticle 2 is projected onto the detection pattern 13 on the detection mask 11). , when the wafer stage 5 is slightly moved, the detection pattern 13 on the detection mask 11 becomes the first detection pattern projection image 14 on the reticle 2.
It moves slightly as shown in Figure 3(a) to Figure 13(b). At this time, while measuring the position of the wafer stage 5 using the length measuring device 15, the projected image 14 of the detection pattern on the reticle 2 is passed through the detection pattern 13, and the light amount is detected by the light amount detector 12. take in. The result is the 14th
As shown in the figure, the amount of light changes in a mountain shape with respect to the position of the wafer stage 5. The position where the amount of light is maximum is the center of the image projected by the detection pattern 9.

Next, as shown in FIG. 11, the wafer stage is moved to the two-dot chain line area, the alignment detector 7 detects the position of the alignment pattern 6 on the light receiving mask 11, and the position of the alignment pattern 6 on the light receiving mask 11 is
From the positions of the alignment pattern 6 and the detection pattern 13 above, find the difference between the detection position of the alignment pattern 6 and the center of the detection pattern projection image 14, and use that value as a correction value to detect the detection pattern by the alignment detector 7. Add to alignment detection position data. That is, the position of the alignment pattern 6 on the wafer 4 was detected by the alignment detector 7, and the reticle pattern projection position was determined by adding a correction value to the detected alignment pattern position data.

[Problem to be solved by the invention]

In the above conventional technology, when detecting the detection pattern on the reticle with exposure light and detecting the position of the image projected onto the detection unit through the reduction lens, the stage is moved to detect the amount of light in order to obtain the correction value. Therefore, not only the correction value detection accuracy of alignment detection deteriorates due to the stage position measurement accuracy and stage mounting accuracy, but also the stage movement and stopping time is required, making it difficult to shorten the detection time.

An object of the present invention is to solve the above-mentioned problems of the prior art, and when detecting a detection pattern on a reticle with exposure light and detecting the position of an image projected onto a detection unit via a reduction lens, stage movement is not required. By not doing so, or by reducing the amount or number of stage movements, not only can errors such as stage position measurement accuracy and stage control accuracy be eliminated or reduced, but also the time required for stage movement and stopping can be eliminated or shortened. By doing so, it is possible to obtain a correction value for alignment detection with high precision and high speed.

This provides a method and apparatus for performing highly accurate alignment.

[Means to solve the problem]

In order to achieve the above objective, in the alignment correction detection system, a plurality of detection patterns on the reticle are arranged at arbitrary intervals, a detection unit consisting of a detection mask and a light intensity detector is provided on the wafer stage, and the detection mask A detection pattern similar to the detection pattern on the reticle is placed on top with a slight deviation from the spacing between the patterns on the reticle,
Further, a light amount detector is provided under each of the plurality of detection patterns, and light amount change data is obtained by detecting the light amount with each light amount detector. The position where the light amount of this light amount change data is maximum is the reticle pattern projection position. Therefore, the reticle pattern projection position can be detected without moving the stage or by reducing the amount or number of movements.

In addition, by making the alignment correction detector a one-dimensional sensor, it is possible to directly detect the detection pattern projection image on the reticle, and the stage can be moved by finding the center of the light amount change data in the same way as above. It is now possible to detect the position of the reticle pattern projection image without having to do so.

[Effect]

When detecting the position of the image projected onto the detection unit by illuminating the detection mark on the reticle with exposure light through the reduction lens, the stage position measurement accuracy is improved by not moving the stage or by reducing the amount or number of movements. This has made it possible to eliminate or reduce error factors such as stage control accuracy, and to eliminate or shorten the time required for stage movement and stopping.

Thereby, the position of the reticle projection image can be determined with high precision and high speed.

Here, the position of the alignment pattern is detected by the alignment detector, and the amount of deviation between the position of the projected image of the reticle pattern and the alignment pattern detection position is determined. By adding this amount of deviation to the position data detected during alignment detection, it becomes possible to relatively determine the position of the projected image of the reticle pattern using the alignment detector.

Therefore, by determining the position of the reticle projected image with high accuracy, the amount of deviation between the alignment pattern detection position and the position of the reticle pattern projected image can be determined with high accuracy.
The alignment detector can determine the position of the projected image of the reticle with high precision.

〔Example〕

The present invention will be explained below using examples.

Embodiment 1 FIG. 3 shows an embodiment in which the present invention is applied to a reduction projection exposure apparatus.

The reduction projection exposure apparatus includes an exposure light source 1, a reticle 2, a reduction lens 3 for reducing and transferring the pattern on the reticle 2,
A wafer stage 5 for adsorbing and moving the wafer 4, a length measuring device 15 for measuring the position of the wafer stage, an alignment detector 7 for detecting the alignment pattern 6 on the wafer 4, and a main unit for controlling the entire apparatus. It mainly consists of a control system 8.

In order to correct the alignment detection here, first,
As shown in FIG. 3(a), by illuminating the detection pattern 9 on the reticle 2 with exposure light, the pattern is projected onto the detection unit 10 provided on the wafer stage 5 via the reduction lens 3. This projected image is detected by the light amount detector 12 in the detection unit 10, and the light amount change signal is taken into the main control system 8 to determine the center position of the pattern projected image. Next, the alignment detector 7 detects the detection unit 1.
Detect the position of the alignment pattern 6 provided on 0,
The position data is taken into the main control system 8. Here, the main control system 8 determines the amount of deviation between the center of this detection pattern projection image and the position of the alignment pattern 6.

At this point, the wafer stage 5 is moved to the position shown in FIG. By adding this correction value to the position data of the alignment mark detected by the alignment detector 7, the position data to which this correction value has been added becomes the position of the projected image of the reticle.

In order to perform the above correction, a plurality of (for example, 5) reticules (for example, 5) must be placed on the reticle 2 at regular intervals (D) as shown in FIG. 1(a).
A correction mark 17 is provided in which detection patterns 9-1 to 9-5 (eg, rectangular openings) are arranged. Further, as shown in FIG. 2, the detection unit 10 provided on the wafer stage is composed of a light receiving mask 11 and a light amount detector 12. As shown in FIG. 1(b), the light-receiving mask 11 has a plurality of pieces (for example:
A correction mark 18 in which five (5) detection patterns 13-1 to 13-5 are arranged is provided, and an alignment pattern 6 is placed next to the detection pattern 13 on the light-receiving mask as shown in FIG. The light amount detector I2 is arranged under each of the detection patterns 13-1 to 13-5 arranged on the light receiving mask 11.

A method for determining the amount of deviation in alignment detection using the correction marks 17 and 18 will be described in detail.

When the correction mark 17 on the reticle is projected onto the surface of the light receiving mask 11, the projected images 14-1 to 14-5 of the detection pattern on the reticle become as indicated by the broken lines shown in FIG. 1(C).

Here, the detection patterns 13-1 to 5 of the light-receiving mask 11
are arranged as shown by the solid line in FIG. 1(C), so when the light receiving mask 11 is removed, the amount of light detected by the light amount detector 12 and taken into the main control system 8 is determined by the fourth It will look like the figure. This is the same result as when the light amount is detected by moving the wafer stage 5 step by step at minute intervals (α) using the conventional method. Therefore, this change in light intensity can be determined by measuring the position of the wafer stage 5 using the length measuring device 15 and importing the position data into the main control system 8.
It is possible to obtain the same light amount change in FIG. 5 as when the detection pattern (eg, 13-3) is moved in steps at intervals of α using an arbitrary detection pattern (eg, 13-3) as a reference. The center position of the detection pattern projection image 14-3 on the reticle 2 is determined from this light amount change, and the center position of the detection pattern 9 on the reticle 2 is illuminated with exposure light and projected onto the detection unit 10 through the reduction lens 3. It will be in the center position.

As a method for determining this center position, there is a threshold method as shown in FIG. 6(a). This is done in advance,
A parameter corresponding to the threshold level R is determined, a threshold is given to the captured data, and the intersection point A is obtained from four or more data points before and after that level.

B can be determined, and the median value of both can be determined as the center position of the projection mark. In addition, as another method, see Figure 6 (
As shown in b), approximation is performed using a quadratic curve using data from several points on the left and right of the maximum value of the captured data, and the position where the maximum value is found is determined as the center position of the projection mark. I can do it.

However, the approximate curve is not necessarily a quadratic curve.

After determining the center position, the alignment detector 7 is used to detect the position of the alignment pattern 6 on the light-receiving mask 11 and input into the main control system 8.

Here, by measuring the position of the wafer stage 5 using the length measuring device 15 and importing it into the main control system 8, the light receiving mask 1
The position of alignment pattern 6 on 1 is determined. the above,
The position of the detection pattern 9 on the reticle 2, which is projected through the reduction lens 3 by illuminating the reticle 2 with exposure light, and the position of the detection pattern 13 and alignment pattern 6 formed on the light-receiving mask 11. Calculate the alignment detection correction value from the positional relationship.

Here, in the above, detection was performed using rectangular openings as the detection pattern 9 on the reticle and the detection pattern 13 on the light-receiving mask, but as shown in FIG. This increases the amount of light and improves detection accuracy. It is also possible to form a slit-like pattern as shown in FIG. 7(b). Further, the structure of the detection unit 10 is as shown in FIG.
8(a), the detection pattern 1 is placed directly on the light receiving surface of the light amount detector 12, as shown in FIG. 8(a).
3 can also be used by providing one or more alignment marks 6 on the detection unit. Furthermore, as shown in FIG. 8(b), it is also possible to use a one-dimensional light amount detector]6 as the light amount detector. It is assumed that the size of the light receiving element in the one-dimensional light intensity detector used here is smaller than the spread of light passing through the detection pattern 13. In order to detect all the amount of light passing through the detection pattern 13, the light amount data of the light receiving elements that are hit by the light coming out from one detection pattern are added together, and the light amount data from that detection pattern is obtained. , the same light amount change data as the light amount detection unit shown in FIG. 2 can be obtained.

As described above, the detection pattern 9 on the reticle 2 is projected through the reduction lens 3 by illuminating the reticle 2 with exposure light.
Since the wafer stage 5 is not moved when detecting the position, it is possible to not only eliminate error factors such as stage position detection accuracy and stage control accuracy, but also to eliminate the time required for stage movement and stopping. becomes.

Example 2 In Example 1, when the number of detection patterns 9 on the reticle and the detection patterns 13 on the light receiving mask is reduced, the amount of detected light data decreases, for example, as shown in FIG. 9(a).
Data such as the initial position shown in is detected.

Here, step movement (for example, 2 step movement between ) The light amount data detected by doing this is the sum of the data after the 1-step movement and the data after the 2-step movement shown in FIG. Data similar to that shown in FIG. 5 is obtained. Therefore, in the same manner as in Example 1, the center position is determined from this light amount change data, and the alignment correction value is determined.

Alternatively, the wafer stage can be moved in the direction of the arrow by an amount (α) that is the distance between the detection patterns 9 placed on the reticle and the distance between the detection patterns 13 placed on the light-receiving mask from the initial position. When scanning is performed, the light amount change data becomes as shown in FIG. 9(b). From this light amount data,
The center position is determined in the same manner as in Example 1, and the alignment correction value is determined.

From the above, in the second embodiment, the number of times or distance of movement is reduced compared to the conventional example, which not only reduces error factors such as stage position detection accuracy and stage control accuracy, but also reduces the amount of time required for stage movement and stopping. It becomes possible to shorten the time.

Example 3 In Example 1, the detection unit IO used both the light receiving mask 11 and the light amount detector 12, but by using a one-dimensional light amount detector, the projected image of the detection pattern on the reticle can be directly detected. Detection becomes possible, and the detected light amount values for each element are as shown in FIG. Here, from this light amount detection value, the center is detected in the same manner as in Example 1, and the position of the reticle projected image is determined. by this,
There is no need to move the stage, which not only eliminates error factors such as stage position detection accuracy and stage control accuracy, but also eliminates the time required to move and stop the stage.

〔Effect of the invention〕

In the conventional method described above, it is difficult to correct the alignment detection system with high precision and at high speed. Therefore, in the alignment correction detection system, multiple detection patterns on the reticle are arranged at arbitrary intervals, and the detection patterns on the wafer stage are arranged with a slight deviation from the intervals at which the detection patterns on the reticle are arranged. By doing so, it is possible to not only eliminate or reduce error factors such as stage position detection accuracy and stage control accuracy by not moving the stage or reducing the amount or number of movements, but also eliminate or reduce the time required for stage movement and stopping. This makes it possible to shorten the length, making it possible to perform alignment correction with high precision and high speed. This makes it possible to improve the accuracy of alignment detection that corrects the error between the position of the reticle pattern projected image and the alignment detection value.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram of the arrangement of detection patterns according to an embodiment of the present invention, FIG. 2 is an explanatory diagram of the configuration of a detection unit, and FIG. 3 is an embodiment of the present invention when installed in a reduction projection exposure apparatus. 4 is an explanatory diagram of the light amount data when the stage is not moved, FIG. 5 is an explanatory diagram of how to handle the light amount data, FIG. 6 is an explanatory diagram of the center position detection method, and FIG. 7 is an explanatory diagram of the detection pattern Fig. 8 is an explanatory diagram of the detection unit, Fig. 9 is an explanatory diagram of light amount data when the amount or number of stage movements is reduced, and Fig. 10 is a diagram showing the direct detection pattern using a one-dimensional light amount detector. Explanatory diagram of light amount data when detected, 11th
The figure is an explanatory diagram of an embodiment of a conventional alignment correction system, FIG. 12 is an explanatory diagram of a detection unit of an embodiment of the conventional system,
FIG. 13 is an explanatory diagram of a detection pattern of a conventional embodiment, and FIG. 14 is an explanatory diagram of light amount data of a conventional embodiment. l... Exposure light source 2... Reticle 3... Reduction lens 4... Wafer 5... Wafer stage 6... Alignment pattern 7... Alignment detector 8... Main control system 9. ...Reticle detection pattern 10...Detection unit 11...Light receiving mask 12
... Light amount detector 13 ... Detection pattern on the light receiving mask 14 ... Detection pattern projection image on the reticle 15 ... Length measuring device
16... One-dimensional light intensity measuring device 17... Correction mark on the reticle 18... Correction mark on the light-receiving mask 19... Representative of the light-receiving element in the one-dimensional detector Patent attorney Katsuo Ogawa Figure i1 (( L) 1'7 (b) LFI (C) Figure 2 Chestnut, Figure 5 (α) 〒4 Figure 5 Wahehihehi Teishi 4F Place Y6 Figure (α) ■ (b) ■ Figure 7 (α) (b) Closed 6 diagram (cL) (b) Death Ω diagram (α) (b) 1; Missing nine, 1-it history, A receiving nine elements A l412 diagram I1j ((L) Cb) χ14 diagram wafer or tray f erection

Claims (1)

[Claims] 1. In a method in which the position of a projected image of a detection pattern on a reticle is detected by a detection pattern and a light amount detector of a detection unit provided on a wafer stage, the pattern on the reticle can be A pattern similar to the detection pattern on the reticle is placed on the wafer stage with a slight shift from the spacing between the patterns on the reticle, and a light intensity detector is placed under each pattern. The position of the projected image of the reticle pattern and the alignment pattern are determined using means for detecting the position of the projected image of the detection pattern on the reticle and means for detecting the position of the alignment pattern provided on the detection unit using an alignment detector. The position of the projected image of the reticle pattern is determined by the alignment detector using means for determining the amount of deviation of the detected position and determining the position of the alignment pattern using the alignment detector, and means for adding the amount of deviation to the detected position data of the alignment pattern. An alignment method characterized by relatively determining the position and reducing alignment errors between the pattern transfer position on the reticle and the circuit pattern on the wafer. 2. In a method in which the position of the projected image of the detection pattern on the reticle is detected by a detection unit installed on the wafer stage, the detection unit is a one-dimensional light amount detector (line sensor) and the position of the projected image of the detection pattern on the reticle is detected directly on the reticle. Using a means for detecting a projected pattern image and a means for detecting the position of the alignment pattern using an alignment detector, the amount of deviation between the position of the reticle pattern projected image and the alignment pattern detection position is determined, and the alignment detector detects the position of the alignment pattern. and adding the amount of deviation to the detected position data of the alignment pattern.
An alignment method characterized by determining the relative position of a projected image of a reticle pattern using an alignment detector to reduce alignment errors between the pattern transfer position of the reticle and the circuit pattern on the wafer.