JP2002198278A - Position measuring equipment and alignment method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide position measuring equipment which can measure a position of a mark included in an image signal accurately in a short time in the case that an obtained image signal is changed in accordance with surface state of a substrate, condition of illumination, etc., and a part of the image signal is lacked, and to provide an alignment method. SOLUTION: When an image signal obtained by photoelectric conversion of an image Im11 or the like which is image-formed on an image sensing surface of an image sensing element is subjected to image-processing, condition of a mark image is predicted on the basis of substrate information (lot information). Consequently, image processing is optimized to substrates in a lot unit even in the case that relation of light and shade in the image Im11 and its peripheral part is changed by the surface state of a substrate and the condition of illumination, and position information (position information of a mark) of the image Im11 is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a position measuring device and an alignment method, and more particularly to a position measuring device and an alignment method suitable for manufacturing a micro device using a large substrate in a lithography process.

[0002]

2. Description of the Related Art Semiconductor devices, imaging devices, liquid crystal display devices,
Alternatively, in the manufacture of a micro device such as a thin film magnetic head, an image of a pattern formed on a mask or a reticle (hereinafter, referred to as a mask) is referred to as a wafer, a glass plate, or the like (hereinafter, these are collectively referred to as a substrate). An exposure device for transferring the image to the substrate is used. For example, a liquid crystal display element (LC
When manufacturing D), a so-called step-and-repeat type exposure apparatus (hereinafter, referred to as a stepper) is often used. The stepper exposes an LCD pattern formed on a mask to a predetermined area of a glass plate (hereinafter, referred to as a plate) as a substrate, and then moves the plate stage on which the plate is mounted by a predetermined distance. Then, another area of the plate is exposed again, and the operation is repeated for all the areas set on the plate, thereby transferring the image of the pattern formed on the mask to the entire plate. .

When transferring an image of a pattern formed on a mask onto a plate, it is necessary to precisely match the position of the plate with the position of the image of the pattern. In recent years, with the miniaturization of the pattern formed on the plate, high alignment accuracy between the mask and the plate is required. Alignment accuracy is required. For this reason, marks for position measurement are formed on the mask and the plate, and the exposure apparatus includes a position measurement device that measures the position information of these marks with high accuracy.

[0004] As a main position measuring device for measuring position information using a position measuring mark formed on a mask or a plate, when classified from the viewpoint of measurement principle, L
SA (Laser Step Alignment) method and FIA (Field Im
age Alignment) method. The LSA type position measurement device is a position measurement device that irradiates a laser beam onto a mark of a dot row formed on a plate and measures position information of the mark by using diffracted / scattered light. F
The IA type position measuring device illuminates a mark using a light source having a wide wavelength band such as a halogen lamp, converts an image of the resulting mark into an image signal, and then performs image processing to obtain position information of the mark. Is a position measuring device for obtaining an asymmetric mark, which is effective for measuring an asymmetric mark formed on an aluminum layer or a plate.

When the position measuring devices are classified from the viewpoint of an observation method in position measurement, a TTL (through-the-lens) for measuring position information of a mark on a plate via a projection optical system is used.
A TTM (through lens) system that simultaneously observes a mark formed on a plate and a mark formed on a mask via a projection optical system and detects a relative positional relationship between the two.
The mask) system, and an off-axis system that directly measures position information of a mark on a plate without using a projection optical system.

A conventional exposure apparatus for manufacturing a liquid crystal display element uses an off-axis type position measuring device as a position measuring device for measuring a position of a plate with respect to a plate stage in order to handle a large plate. Further, the above-mentioned TTL position measuring device is provided for measuring the position of the mask with respect to the plate stage. All of the position measuring devices provided in these conventional exposure apparatuses determine the position of the plate stage by measuring the position of a reference member formed on the stage.

An exposure apparatus having an off-axis type position measuring device for measuring plate position information has a measurement field of view set outside a projection area of a projection optical system that irradiates a plate with an image of a pattern formed on a mask. Off
A reference position set in the axis measuring device,
A baseline amount, which is the distance from the reference position of the projected image of the pattern formed on the mask, is obtained in advance. This baseline amount is obtained from the result of measuring the above-described reference member with each position measuring device.

[0008]

Recently, a position measuring device for measuring the position of a plate and a position measuring device for measuring a relative position between a mask and a plate are of the FIA type mainly from the viewpoint of shortening the measuring time. Tend to be frequently used. The FIA type position measurement device performs image processing on an image signal obtained by photoelectrically converting an optical image of a measurement target (for example, a mark formed on a plate) formed on an imaging surface of an imaging element, The position of the measurement target is obtained by calculating the reference position of the measurement target (for example, the center position of the mark) and calculating the amount of deviation between this reference position and the reference position (for example, the center of the measurement visual field) set in the position measurement device. Ask for.

The processing algorithm of the image processing applied to the image signal is predetermined, and the image processing is performed according to the processing algorithm. The image signal is a signal whose signal intensity changes in accordance with, for example, the shape of the mark formed on the plate, and the like. Find the center position of the mark.

Here, for example, consider the case of measuring the position of a mark formed on a plate. The state of the surface of the plate changes variously depending on the process. Therefore, even when the mark is designed so that the signal strength of the part where the mark element is formed is higher than the signal strength obtained from the surrounding area, the signal strength of the part where the mark element is formed is conversely increased. In many cases, the signal strength is lower than the signal strength in the peripheral part. Such a situation depends not only on the surface condition of the plate but also on the method of forming the mark, the illumination condition of the mark, and the like.

In the above example, the processing algorithm of the above-described image processing is set so as to detect the center position of the mark on the assumption that the signal strength of the portion where the mark element is formed is higher than that of the peripheral portion. I have. Therefore, if the signal intensity of the portion where the mark element is formed is lower than that of the peripheral portion due to the surface condition of the substrate or the illumination condition, the center position of the mark cannot be detected and an error occurs. Every time an error occurs, the processing is interrupted, and the throughput (the number of plates that can perform the exposure processing per unit time) is reduced. Usually, a plurality of types of image processing algorithms are prepared in advance in order to prevent this decrease in throughput. In the above example, a processing algorithm for measuring the center position of the mark is prepared when the signal strength of the peripheral part is high and the signal strength of the part where the mark element is formed is low.

However, it is impossible to know whether the signal intensity of the portion where the mark element is formed or the signal intensity of the peripheral portion becomes higher until the measurement is actually performed. Therefore, conventionally, for example, image processing is performed using a processing algorithm in a case where the signal intensity of a portion where the mark element is formed is higher than that of the peripheral portion, and when an error occurs in this processing, the mark element is more than the peripheral portion. Image processing is performed using a processing algorithm when the signal strength of the formed portion is low. With such processing, re-measurement is automatically performed, so that a drop in throughput due to the interruption of the processing described above can be prevented. However, when performing the above-described re-measurement, since the surface conditions in the same lot or related lots are all performed on almost the same substrate, conventionally, useless processing is repeatedly performed. At present, there is an increasing demand for improving the throughput mainly to reduce the manufacturing cost, and it is necessary to reduce such useless processing as much as possible.

Further, in the FIA type position measuring apparatus which is frequently used at present, the size of the measurement visual field is limited by the magnification of the optical system or the like, and the mark or the like to be measured is accurately positioned within the measurement visual field. If it is not arranged, the position of the mark cannot be measured. Usually, the position of the plate is controlled based on the process data storing the information on the position of the mark and the like, so that the mark to be measured is almost accurately arranged in the measurement visual field. However, it is conceivable that the entire mark is not arranged in the measurement visual field due to a control error of the stage for moving the plate while the plate is placed, and a situation where a part of the mark is missing occurs. In this case, the position of the mark cannot be detected, an error occurs, and the processing is interrupted. In order to restart the interrupted process, it is necessary to perform a manual assist process and restart the process, which causes a problem of lowering the throughput.

In particular, in an exposure apparatus that exposes a large-area plate, the process of transferring the image of the pattern formed on the mask to the plate itself requires time, and therefore the number of plates that can be processed per unit time is small. (Low throughput). Therefore, it is extremely important to omit wasteful processing and increase the throughput in order to increase the manufacturing efficiency and reduce the manufacturing cost. Since it is quite difficult to reduce the time required to transfer the pattern image to the plate, it is important to reduce the time required for mark measurement other than the time required for transfer.

The present invention has been made in view of the above circumstances, and is intended for a case where an image signal obtained according to the surface condition of a substrate or an illumination condition changes or a case where a part of the image signal is missing. It is another object of the present invention to provide a position measuring device and an alignment method that can accurately and quickly measure the position of a mark included in an image signal.

[0016]

In order to solve the above-mentioned problems, a position measuring apparatus according to a first aspect of the present invention comprises a mark provided for each substrate (P) on a substrate (P) in lot units. The optical system (34, 3) that projects (AM1 to AM4)
5, 36, 38, 39, 40, 41, 42) and the optical system (34, 35, 36, 38, 39, 40, 41, 4).
2) an image sensor (42) for outputting an image signal obtained by photoelectrically converting the optical image projected through the image sensor (2), and an image signal output from the image sensor (42) is processed to form a mark included in the optical image. A processing unit (15) for obtaining position information of an image, wherein the processing unit (15)
Is characterized in that, based on the substrate information of the substrate (P) in the lot unit, a state of a mark image of the image signal is predicted and subjected to image processing to obtain position information of the marks (AM1 to AM4). . According to the present invention, when performing image processing on an image signal output from an image sensor, position information of a mark image provided on a substrate is obtained by performing image processing by predicting a state of a mark image based on substrate information. Is required. Therefore, image processing is optimized for a lot-by-lot substrate, and it is possible to eliminate or minimize the situation where mark position information is not required, so that the position of a mark can be measured in a short time. As a result, the throughput can be improved. Here, the processing means (15) has a plurality of image processing algorithms for detecting an edge portion of the mark (AM1 to AM4), and the processing means (15) according to the density relationship of the mark (AM1 to AM4) image of the image signal. Preferably, the plurality of image processing algorithms are switched. A storage unit (15a) for storing image processing information in the image processing of the image signal and the substrate (P) information for each lot; and wherein the processing unit (15) is provided in the storage unit (15a). Preferably, one of the plurality of image processing algorithms is prioritized based on the stored image processing information. Further, it is preferable that the storage means (15a) stores image processing information for each substrate (P) in lot units for which position measurement is performed. Further, it is preferable that the image signal includes a plurality of mark images, and the storage means (15a) stores image processing information for each mark image. The optical system (34, 35, 36, 38, 39, 40, 41, 4)
2) are marks (AM1 to AM1) provided on the substrate (P).
AM4) In addition to projecting an index serving as a measurement reference in the vicinity,
The projected mark image and the marks (AM1 to AM4)
Is projected on the detection surface of the image sensor (42). In order to solve the above-mentioned problems, the second aspect of the present invention
The position measuring device according to the aspect of the present invention includes an optical system (34, 3) for projecting marks (AM1 to AM4) provided on
5, 36, 38, 39, 40, 41, 42) and the optical system (34, 35, 36, 38, 39, 40, 41, 4).
2) an image sensor (42) for outputting an image signal obtained by photoelectrically converting the optical image projected through the image sensor (2), and an image signal output from the image sensor (42) is processed to form a mark included in the optical image. In a position measuring device provided with processing means (15) for obtaining position information of an image, the reference position (G1, G1)
First template provided with 1) (55a, 58a)
And the first template (55a, 58a) and the reference position (G1, G1) with respect to a predetermined axis (56, 59).
The second template (55) formed in a mirror image relationship with (1)
b, 58b) and the first template (55a, 58b) for each of the symmetric positions of the mark image.
a) and the second template (55b, 58b) are subjected to pattern matching, and the positions of the marks (AM1 to AM4) from the reference positions (G1, G2, G11, G12) provided in the respective templates are determined. And a required image processing unit (15). According to the present invention, template matching is performed on an image signal using a first template and a second template that are mirror images of each other with respect to a predetermined axis, and mark position information is obtained from each reference position. Therefore, the position of the mark can be measured accurately and in a short time. In order to solve the above problem, an alignment method according to a first aspect of the present invention is directed to an alignment method for detecting a position of an alignment mark (IM) provided on a substrate (M).
A plurality of mark elements (Pa1 to Pa5,
Pb1 to Pb5, Pc1 to Pc5, and Pd1 to Pd5) are arranged symmetrically and the alignment mark (I
M), and a second step of obtaining an approximate center position of the imaged alignment mark (IM).
A step of selecting a symmetrically arranged pair of mark elements based on the center position obtained in the second step; and a signal processing of the selected pair of mark elements, Alignment mark (IM)
And a fourth step of obtaining the center position of According to the present invention, first, an approximate center position of the alignment mark is obtained, a pair of symmetrically arranged mark elements is selected, and signal processing is performed on the selected pair of mark elements to obtain position information of the alignment mark. Because
For example, even when an image signal in which a part of the alignment mark is missing is obtained, the position information of the alignment mark can be obtained from the remaining mark elements constituting the alignment mark without significantly deteriorating the accuracy.
In addition, the fourth step is a step in which the symmetrically arranged mark elements (Pa1 to Pa5, Pb1 to Pb5, Pc1 to Pc
For the pair of c5, Pd1 to Pd5), it is preferable to perform signal processing by weighting the pair of mark elements closer to the center of the alignment mark (IM). Further, the alignment mark (IM) is an alignment mark in which a plurality of pairs of mark elements are arranged symmetrically with respect to a direction orthogonal to the predetermined axis, and the center positions in the respective directions are substantially the same. It is preferable to determine the center position of the alignment mark by performing the third step and the fourth step in the direction. In order to solve the above-mentioned problem, an alignment method according to a second aspect of the present invention is a method of aligning alignment marks (AM1 to AM4) provided for each substrate (P) with respect to a substrate (P) in lot units. In the alignment method for detecting, an alignment mark (AM) on the substrate (P) is used.
1 to AM4), and stores a processing algorithm used at the time of detecting the position of the imaged alignment mark (AM1 to AM4). According to the present invention, since the processing algorithm used at the time of position detection by imaging the alignment mark is stored, when processing another substrate in the same lot or a related lot, the processing algorithm is used. If the processing is performed using, the situation that the position information of the alignment mark cannot be obtained can be eliminated or extremely reduced.
The position information of the alignment mark can be obtained in a short time.
The alignment method according to the second aspect of the present invention includes:
In the alignment of the substrate (P) in the lot unit,
It is characterized in that position detection is performed using a processing algorithm for successful position detection on the substrate (P) of the same lot or a related lot.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a position measuring device and an alignment method according to an embodiment of the present invention will be described in detail with reference to the drawings. [Structure of Exposure Apparatus] FIG. 1 is a perspective view showing a schematic configuration of an exposure apparatus provided with a position measuring apparatus according to an embodiment of the present invention, and FIG. 2 is provided with a position measuring apparatus according to an embodiment of the present invention. FIG. 2 is a front view illustrating a schematic configuration of the exposure apparatus. In the present embodiment, an example is described in which the present invention is applied to an exposure apparatus that transfers an image of the pattern to a plate P by a step-and-repeat method using a plurality of masks on which patterns of a liquid crystal display element are formed. Will be explained. still,
In the following description, the XYZ rectangular coordinate system shown in FIG. 1 is set, and the positional relationship of each member will be described with reference to the XYZ rectangular coordinate system. The XYZ rectangular coordinate system is
The X axis and the Y axis are set to be parallel to the plate stage PS, and the Z axis is set to the plate stage PS
Are set in a direction orthogonal to the direction (a direction parallel to the optical axis of the projection optical system PL). In the XYZ coordinate system in the figure, the XY plane is actually set as a plane parallel to the horizontal plane, and the Z axis is set vertically upward.

A light source 2 such as an ultra-high pressure mercury lamp for supplying exposure light (illumination light) for g-line (436 nm) and i-line (365 nm) is arranged at a first focal position of the elliptical mirror 1. The exposure light from the light source 2 is condensed by the elliptical mirror 1 and reflected by the mirror 3, then converged at the second focal position of the elliptical mirror 1 and enters the wavelength selection filter 4. The wavelength selection filter 4 has a wavelength (g line, h line, i
Line) only. The illumination light having passed through the wavelength selection filter 4 is converted into a light flux having a uniform illuminance distribution by a fly-eye integrator 6.

Between the wavelength selection filter 4 and the fly-eye integrator 6, a dimming filter 5 is disposed so as to be able to advance and retreat with respect to the optical path. The neutral density filter 5 suppresses the amount of exposure light that has passed through the wavelength selection filter 4.
The fly-eye integrator 6 is composed of a large number of positive lens elements, and forms, on its exit side, light source images of the same number of positive lens elements to form a substantial surface light source. Although not shown in FIG. 1, the exit surface of the fly-eye integrator 6 has a σ value for determining the illumination condition (on the pupil plane with respect to the aperture diameter of the pupil of the projection optical system PL described later). A diaphragm member for setting the ratio of the aperture of the light source image at the aperture is provided.

Light beams from a large number of light source images formed by the fly-eye integrator 6 irradiate a blind 7 for increasing or decreasing the size of the aperture S to adjust the irradiation range of exposure light. The exposure light that has passed through the opening S of the blind 7 is reflected by the mirror 8 and enters the condenser lens 9, and the condenser lens 9 forms an image of the opening S of the blind 7 on the mask M placed on the mask stage MS. And the desired area of the mask M is illuminated. The replacement of the mask M placed on the mask stage MS is performed by a mask changer (not shown). Mask changers are disclosed in, for example, Japanese Patent Application Laid-Open Nos. 6-310398 and 9-283.
It is preferable to use a reticle changer disclosed in Japanese Patent Application Laid-Open No. 416, JP-A-11-204431, or the like. The elliptical mirror 1, the light source 2, the mirror 3, the wavelength selection filter 4, the neutral density filter 5, the fly-eye integrator 6, the blind 7, the mirror 8, and the condenser lens 9 constitute an illumination optical system.

An image of the device pattern DP existing in the illumination area of the mask M is provided on a plate P (see FIG. 2) as a substrate mounted on a plate holder 11 provided on a plate stage PS via a projection optical system PL. ). The projection optical system PL is provided with a lens controller unit 10 that controls optical characteristics such as imaging characteristics to be constant in response to environmental changes such as temperature and atmospheric pressure. In FIG. 2, the upper surface of the plate P is shown by a broken line, but this shows the upper surface when the plate P is mounted on the plate stage PS. The plate stage PS is obtained by superimposing a pair of blocks PS _X and PS _Y that can be moved in the X-axis direction and the Y-axis direction in the figure, and the position in the XY plane can be adjusted. , In FIG.
For convenience blocks S _X, it is illustrated as being integrally PS _Y). Although not shown, the plate stage PS includes a Z stage for moving the plate P in the Z axis direction, a stage for slightly rotating the plate P in the XY plane, and an XY plane by changing the angle with respect to the Z axis. And the like, which adjusts the inclination of the plate P with respect to. One end of the upper surface of the plate holder 11 has a movable mirror 12 having a length longer than the movable range of the plate stage PS.
X and 12Y are attached, and laser interferometers 13X and 13Y are arranged at positions facing the mirror surfaces of the movable mirrors 12X and 12Y, respectively.

1 and 2, the laser interferometer 13X is provided with a movable mirror 1 along the X-axis.
Two X-axis laser interferometers for irradiating a 2X laser beam are provided, and one X-axis laser interferometer and one laser interferometer 13Y are used to set the X coordinate and Y of the plate holder 11.
The coordinates are measured. The rotation angle of the plate holder 11 in the XY plane is measured based on the difference between the measurement values of the two X-axis laser interferometers. Laser interferometer 13X, 1
Information on the X coordinate, the Y coordinate, and the rotation angle measured by 3Y is supplied to the main control system 15 as stage position information. The main control system 15 outputs the supplied stage position information to the stage drive systems 14X and 14Y while monitoring the supplied stage position information, and controls the positioning operation of the plate stage PS.

Here, the arrangement of members provided on the plate holder 11 for measuring various types of positional information will be described. FIG. 3 is a diagram showing the arrangement of members provided on the plate holder 11 used when measuring various types of position information. As shown in FIG. 3, the plate holder 11 is provided with a reference member FM, reflectors RF1 to RF4, and detection holes AS1 and AS2. The reference member FM and the reflection plates RF <b> 1 to RF <b> 4 are configured to be freely protruded and immersed in the plate holder 11. It is designed to be. By immersing them in the plate stage PS, the plate P can be placed on the plate holder 11. The reason that the reference member FM and the reflection plates RF1 to RF4 can be freely protruded and immersed is to reduce the size of the plate stage PS. In other words, if these are formed other than the portion where the plate P is placed,
This is because the size of the plate stage PS increases accordingly. In addition, what kind of position information the reference member FM, the reflection plates RF1 to RF4, and the detection holes AS1 and AS2 are used for measuring will be described below as needed. In FIG. 3, the plate alignment sensors 18a to 18a shown in FIG.
The arrangement relation of 18d is also shown. Plate stage P
When S moves in the XY plane, the reference member FM and the reflection plate RF1 are transmitted to the plate alignment sensors 18a to 18d.
To RF4 and the detection holes AS1 and AS2 move.

Returning to FIG. 1, the position information of the mask M is measured by mask observation systems 16a and 16b provided above the mask M. The mask observation systems 16a and 16b
The detection light is applied to the position measurement mark RM drawn outside the area where the device pattern DP is formed, and the reflected light is received to measure the position of the position measurement mark RM. And the measurement results are sent to the main control system 1
5 is output. The main control system 15 includes a mask observation system 16a,
Based on the measurement result of 16b, the mask stage MS (see FIG. 2) holding the mask M can be moved to a desired position on the XY plane by servo-controlling the driving means 17 such as a linear motor. It has a configuration.
Similarly, above the mask M, an alignment system (not shown) for detecting an image processing mark IM of the mask M, which will be described later, by a TTL method via a reference member FM is provided.

Here, an example of the image processing mark IM as an alignment mark formed on the mask M will be described. FIG. 4 is a diagram illustrating an example of the shape of the image processing mark IM formed on the mask M. Image processing mark I
M processes only the simultaneous measurement marks MA and MB used for simultaneously processing position information in two directions of the X-axis direction and the Y-axis direction, and only one direction in the X-axis direction or the Y-axis direction. The unidirectional measurement marks MC, MD, etc. used for this purpose are set and can be applied according to the situation. The simultaneous measurement mark MA has a shape in which so-called line and space marks are arranged in the X-axis direction and the Y-axis direction.
It has a cross shape in which a linear pattern set in the axial direction and a linear pattern whose longitudinal direction is set in the X-axis direction intersect. In addition, the unidirectional measurement mark MC is
And the space mark is arranged in the X-axis direction, and the unidirectional measurement mark MD has the line and
This is a shape in which space marks are arranged in the Y-axis direction. Here, for example, the image processing mark MA is formed by arranging a plurality of mark elements symmetrically with respect to a predetermined axis. For example, the mark element Pa1 and the mark element Pb1 form a pair, and the mark element Pc1. And the mark element Pd1 are formed as a pair.

Note that the simultaneous measurement marks MA and MB and the edge direction measurement marks MC and MD, which are given as examples of the image processing mark IM, correspond to the measurement field of view VF of the alignment system and the aerial image detection devices 20a and 20b described later. The mark line width and the number of marks can be arbitrarily set as long as they fall within the range. Normal,
Since the device pattern DP, the mark RM for position measurement, and the mark IM for image processing have drawing errors, and depending on the position of the mask M, it is necessary to consider the influence of the distortion of the projection optical system PL. Mark RM and image processing mark IM are device patterns D
P is arranged in the vicinity of the drawn area.

On the other hand, the position information of the plate stage PS in the XY plane is obtained from the plate alignment sensors 18a to 18a.
The position in the Z-axis direction is measured by using an autofocus mechanism 19 including a light projecting system 19a and a light receiving system 19b disposed above the plate stage PS and beside the projection optical system PL. Measured. The auto focus mechanism 19 converts the detection light emitted from the light projecting system 19a into a reference member FM provided on the plate stage PS.
(See FIG. 3) is irradiated from an oblique direction, and the reflected light is received by a light receiving system 19b. The alignment is performed at a position optically conjugate with the mask M mounted thereon. Although not shown in FIG. 1, at least two of the reference member FM and the reflectors RF2 to RF4 are sequentially irradiated with detection light from an oblique direction, and the Z of the plate stage PS is illuminated.
A measuring device for measuring the position information in the axial direction and the inclination of the plate stage PS is also provided.

Here, the plate alignment sensor 18
The configuration of a to 18d will be described. FIG. 5 is a diagram showing a configuration of an optical system of the plate alignment sensors 18a to 18d. The plate alignment sensors 18a to 18a
18d are the same, so in FIG.
Only the configuration of the plate alignment sensor 18a is shown as a representative. In FIG. 5, 30 is 400-800.
This is a halogen lamp that emits light having a wavelength bandwidth of about nm. The light emitted from the halogen lamp 30 is
After being converted into parallel light by the condenser lens 31,
The light enters the dichroic filter 32. The dichroic filter 32 is composed of a plurality of filters configured to freely enter and exit the optical path of the light emitted from the halogen lamp 30. By changing the combination of filters inserted into the optical path, a predetermined number of incident Only the light in the wavelength band is selected and transmitted. For example, of the light emitted from the halogen lamp 30, light in a blue region (wavelength region:
About 420-530 nm), light in the red region (wavelength region: about 580-730 nm), only light in the yellow region,
Alternatively, white light (almost the entire wavelength band of light emitted from the halogen lamp 30) can be transmitted.

The light transmitted through the dichroic filter 32 is incident on a condenser lens 33 whose rear focal point is set substantially at the position of the incident end 34a of the optical fiber 34. The optical fiber 34 has one input end and four output ends, each of which is a plate alignment sensor 18.
a to 18d. The halogen lamp 30, the condenser lens 31, the dichroic filter 32, and the condenser lens 33 are provided outside the chamber of the exposure apparatus in order to avoid a thermal influence, and the light transmitted through the dichroic filter 32 It is configured to guide each of the alignment sensors 18a to 18d.

Light emitted from one emission end 34b of the optical fiber 34 is used as detection light IL1. The detection light IL1 illuminates an index plate 36 on which an index mark 37 of a predetermined shape is formed via a condenser lens 35. Here, the configuration of the index plate 36 will be described. FIG. 6 is a diagram illustrating an example of the shape of the index mark 37 formed on the index plate 36. The index mark 37 includes a first light-shielding portion 37a for shaping the incident detection light IL1 into a rectangular shape and a second light-shielding portion 37b for shading a part of the detection light IL1 shaped into a rectangular shape into an annular shape having four sides. Are formed. The light-shielding portion 37a applies the detection light I to the mark formed on the plate P.
It is formed to define an irradiation area when irradiating L1. The second light-shielding portion 37b is formed to determine a reference position when measuring position information of a mark to be measured. That is, the index mark 37 serves both as a field stop and a mark that determines a reference position used when measuring mark position information.

FIG. 7 is an enlarged view of the second light shielding portion 37b. As shown in FIG. 7, the second light-shielding portion 37b is formed in an annular shape having four sides, and cross-shaped light-shielding portions q1 to q4 are formed at four corners. The light shielding portions q1 to q4 are
It is used when adjusting the arrangement of the index plates 36 provided in a to 18d. Adjustment of the arrangement of the index plate 36 is performed by the second light shielding unit 3.
The image 7b is picked up by an image pickup device 42 (see FIG. 5) described later, and the image is picked up while observing the position of the image in a monitor (not shown). At this time, a linear test pattern is displayed in the scanning line direction of the monitor or in a direction perpendicular to the scanning line, and adjustment is performed so that this test pattern and each side of the second light shielding portion 37b are parallel. I do.

Normally, since the test pattern is narrower than the width of each side of the second light-shielding portion 37b, variations may occur when each side of the second light-shielding portion 37b is directly parallel to the test pattern. . Therefore, cross-shaped light shielding portions q1 to q4 shown in FIG. 7 are provided at the four corners of the second light shielding portion 37b.
The test pattern displayed on the monitor is adjusted so as to pass through at least two adjacent cross-shaped light-shielding portions q1 to q4. The cross-shaped light-shielding portions q1 to q4 are set to have a width of about one third of each side of the second light-shielding portion 37b, so that the accuracy of the adjustment is improved and the adjustment can be easily performed. The index plate 36 on which the index mark 37 shown in FIG. 5 is formed is arranged at a position that is optically conjugate with the imaging plane FC of the detection light IL. The mark formed on the plate P is arranged on the image forming surface FC, and the second light shielding portion 3 is placed on the mark.
The detection light IL including the image 7b is irradiated. When the focal plane of the aerial image detection device 20a is disposed on the imaging plane FC, the aerial image of the image of the second light shielding unit 37b is detected. In the following description, the image of the second light-shielding portion 37b is referred to as an image of an index mark for simplification of the description.

Returning to FIG. 5, the detection light I passing through the reference plate 36
L1 is incident on a half mirror 39 via a relay lens 38 for splitting light transmission and light reception. The detection light IL <b> 1 reflected by the half mirror 39 forms an image on an imaging plane FC via the objective lens 40. When the mark formed on the plate P is arranged on the imaging plane FC, the reflected light passes through the objective lens 40, the half mirror 39, and the second objective lens 41 in order of the imaging element 42 including a CCD or the like. An image is formed on the imaging surface. The optical fiber 34, condenser lens 35, index plate 36, relay lens 38, half mirror 39, objective lens 40, and second objective lens 41 described above constitute an optical system according to the present invention.

The image pickup element 42 sequentially converts the optical images formed on the respective image pickup surfaces into image signals while scanning them in a predetermined scanning direction, and outputs the image signals to the main control system 15 as processing means. The main control system 15 performs image processing of the image signal output from the image sensor 42, and obtains, for example, the center position of the index mark from the center position of the imaging surface of the image sensor 42 and the amount of deviation of the center position of the mark. The details of the image processing performed by the main control system 15 on the image signal output from the image sensor 42 will be described later. Since the laser interferometers 13X and 13Y constantly input stage position information indicating the position of the plate holder 11 in the X-axis direction and the Y-axis direction and the amount of rotation of the plate holder 11, the main control system 15 Reference numeral 15 determines the position information of the mark being measured based on the stage position information and the center position of the index mark from the center position of the imaging surface of the image sensor 42 and the amount of deviation of the center position of the mark.

Here, the plate alignment sensor 18
Since the size of the measurement field of view a to 18d is limited to some extent, the AM1 to AM1 formed on the plate P when performing the measurement
AM4 is accurately detected by the plate alignment sensor 18a-1.
It is often difficult to arrange at the measurement center of 8d. Therefore, in the present embodiment, first, rough position information of the mark is obtained, the plate P is moved based on the position information, and the mark is arranged at the measurement center of the plate alignment sensors 18a to 18d. Seeking. Conventionally, plate alignment sensors equipped with an image sensor measure the approximate position information of the mark by setting the optical magnification of the light-receiving optical system low, and then set the optical magnification to a high value to achieve high-precision position information. However, there is a problem in that it takes time to switch the magnification and the throughput is reduced. The plate alignment sensors 18a to 18b provided in the present embodiment are used to measure rough position information (hereinafter, referred to as search measurement) and to measure position information with high accuracy (hereinafter, referred to as fine measurement). The throughput is prevented from lowering by devising the measurement process by providing different measurement areas on the imaging surface of the image sensor 42.

FIG. 8 is a diagram showing a measurement area for search measurement and a measurement area for fine measurement. In FIG. 8, F
P indicates an imaging surface of the imaging element 42, and the imaging element 42
Is set, for example, in the X-axis direction. As shown, a measurement area SC1 for fine measurement in which the longitudinal direction is set in the X-axis direction including the center of the imaging plane FP, and the longitudinal direction is set in the Y-axis direction including the center of the imaging plane FP. The measurement area SC2 for the fine measurement is set. Further, around the imaging plane FP, measurement areas W11 and W12 and measurement areas W21 and W21, which are measurement areas for search measurement, are arranged.
W22 is set. The measurement regions W11 and W12 have their longitudinal directions set in the X-axis direction, and the measurement regions W2 and W22 have their longitudinal directions set in the Y-axis direction. In the case of search measurement, the main control system 15 performs image processing of an image signal output from the image sensor 42 to measure the measurement areas W11, W12, and W2.
1, the end (edge position) of the mark image formed on W22
Is detected, the center position of the mark is calculated, and the shift amount from the center position of the imaging plane FP is obtained. In the case of fine measurement, the center position of the mark is calculated by detecting the end (edge position) of the image of the mark formed in the measurement areas SC1 and SC2, and the amount of deviation from the center position of the imaging plane FP is calculated. Ask for.

Here, the mark A formed on the plate P
The shape of M will be described. FIG. 9 is a diagram showing the shape of the mark AM formed on the plate P. As shown in FIG. 9, the mark AM has a U-shaped mark element 50a and a U-shaped mark element 50b arranged in the X-axis direction, and has a U-shaped mark element 50c and a U-shaped mark. The elements 50d are arranged in the Y-axis direction. Here, the opening 51a of the U-shaped mark element 50a and the opening 51b of the U-shaped mark element 50b are set to be opposite to each other, and the U-shaped mark element 50c is set.
And the opening 51d of the U-shaped mark element 50d are set to be opposite to each other. Further, in the mark AM, an L-shaped mark element 52a whose longitudinal direction is set in each of the X-axis direction and the Y-axis direction is disposed adjacent to the U-shaped mark elements 50a and 50c. The shape mark element 52b is replaced with the mark element 50b.
Is arranged adjacent to the mark element 50c, the L-shaped mark element 52c is arranged adjacent to the mark element 50b and the mark element 50d, and the L-shaped mark element 52d is adjacent to the mark element 50c and the mark element 50d. It is arranged. The mark AM indicates the measurement areas W11, W12,
When measuring the position information using the image sensor 42 on which W21 and W22 are formed, a shape suitable for measuring the position information in the X-axis direction and the position information in the Y-axis direction without erroneous measurement and at high speed. is there.

Referring back to FIGS. 1 and 2, the exposure apparatus according to the present embodiment includes an aerial image projected via the projection optical system PL,
Alternatively, the above-described plate alignment sensors 18a to 18
The space holders 20a and 20b for detecting the images of the index marks emitted from the respective parts d are provided in the wafer holder 11. The aerial image detecting device 20a has the detection hole AS shown in FIG.
1 is formed at a position where the aerial image detection device 20b is formed.
Is provided at the position where the detection hole AS2 shown in FIG. 3 is formed. FIG. 10 is a cross-sectional view illustrating the configuration of the aerial image detection devices 20a and 20b. Since the configurations of the aerial image detection devices 20a and 20b are the same, only the aerial image detection device 20a is shown in FIG. As shown in FIG.
The aerial image detection device 22a includes a detection hole AS1 formed on the upper surface of the stage holder 11 and having an inner diameter of about several millimeters, a first relay lens 21, a second relay lens 22, and a CC.
D and the like.

The first relay lens 21 and the second relay lens 22 are composed of a focal plane PP1 and an image sensor 23 which are set so as to be substantially the same as the upper surface of the plate P when the plate P is mounted on the stage PS. Is made optically conjugate with the imaging surface. Therefore, when the focal plane PP1 coincides with the imaging plane of the projection optical system PL, the position measurement mark RM or the image processing mark I projected via the projection optical system PL is projected.
The contrast of the aerial image of M becomes sharpest, and the focal plane P
When P1 coincides with the image planes FC (see FIG. 5) of the plate alignment sensors 18a to 18d, the contrast of the image of the index mark becomes clearest. Image sensor 23
The image signal of the aerial image captured at is output to the main control system 15 and subjected to various image processing described later. The reason why the two aerial image detection devices 20a and 20b are provided is to reduce the amount of movement of the plate stage PS to prevent a decrease in throughput.

Further, as shown in FIG. 2, in the present embodiment, a storage device 15a as storage means is connected to the main control system 15. The storage device 15a stores information on the plate P to be processed (including position measurement processing and exposure processing) and image processing information for each substrate, for each mark formed on the substrate, or for each lot. Here, the information on the plate P is process data, for example, information indicating the type of substrate, information indicating what kind of processing is performed on the surface, information indicating a mark forming position, and a position in a lot. And other information. Further, the image processing information refers to the plate alignment sensors 20a to 20a-1.
8d, the image sensor 42 and the aerial image detection device 20a,
The information indicates the processing algorithm used at the time of image processing among the plurality of processing algorithms provided for the image signal output from the image sensor 23 included in the image sensor 20b. The main control system 15 transmits the image data from the image sensor 42 included in the plate alignment sensors 20a to 18d and the image sensor 23 included in the spatial image detectors 20a and 20b according to the information on the plate P and the image processing information stored in the storage device 15a. A processing algorithm for image processing to be applied to the output image signal is determined, and image processing is performed using the determined processing algorithm to obtain mark position information.

[Processing Algorithms Provided in Main Control System] Next, processing contents of a plurality of processing algorithms provided in the main control system 15 will be described. [Detection of Midpoint Between Differential Signal Peaks] FIG. 11 shows an image processing mark IM using the aerial image detection devices 20a and 20b.
FIG. 9 is a diagram for explaining the processing of position information measurement of FIG. FIG.
In FIG. 1, a description will be given of an example in which the XY simultaneous measurement mark MA shown in FIG. 4 is used as the image processing mark IM. First, as shown in FIG.
A cross-shaped image processing range VA is previously arranged and set with reference to the center CP1 of the measurement visual field VF of the image sensor 23. As shown in FIG. 11B, when the image of the mark IM for image processing formed on the mask M is formed on the image sensor 23, the X-axis direction and the Y direction are measured within the measurement visual field VF of the image sensor 23. From the processing ranges VA in the two axial directions, the signal intensities of the image processing marks IM are detected by the designated number, respectively. Then, as shown in FIGS. 12A and 12B, a differentiation process is performed on the raw signal of the mark image captured in the processing range VA to obtain a signal having an intensity as shown in FIG. 12C. . FIG. 12 is a diagram for explaining the differential processing performed on the image signal. FIG. 12C is a diagram for explaining a process of performing a differentiation process on the image signal and obtaining the position information of the image processing mark IM from the midpoint of the edge position of the image processing mark IM. Thereafter, the coordinate position of the center CP4 of the mark is detected by determining the center of each line based on the peak of the obtained differential signal and the midpoint of the peak. By measuring the difference between the mark center CP4 and the visual field center CP1 of the image sensor 23, the coordinate position of the mark center CP4 in the coordinate system of the plate holder 11 is detected as in the case of the pattern matching.

[Slice of Image Intensity Distribution by Raw Signal] An example in which a mark MA for simultaneous XY measurement shown in FIG. 4 is used as the mark IM for image processing will be described. First, as shown in FIGS. 11A and 11B, the image processing range VA is set on the basis of the center CP1 of the measurement visual field VF of the image sensor 23 as in the case of detecting the midpoint of the differential signal. In addition, when the mark M on the mask M is formed on the image sensor 23, the mark M is specified from the processing ranges VA in the X-axis direction and the Y-axis direction in the measurement visual field VF of the image sensor 23, respectively. The signal intensity of the mark M is detected for each number.
Then, as shown in FIGS. 13A and 13B, the raw signal of the image of the mark M captured in the processing range VA is arbitrarily positioned from the bottom for each line (for example, 70% of the total height).
%), And the center point at the sliced point is detected. FIG. 13 is a diagram for explaining slice processing performed on an image signal. In this case, slices may be performed at a plurality of positions, and an average value of the detected center points may be used.
Thereafter, the center C of the mark is shifted from the center point of the detected line.
The coordinate position of P4 is detected. Then, by performing the same procedure as that for detecting the midpoint of the differential signal, the coordinate position of the mark center CP4 in the coordinate system of the plate holder 11 is detected.

[Pattern Matching by Correlation Method (Template Matching)] The mark I for image processing is again used here.
An example in which the XY simultaneous measurement mark MA shown in FIG. 4 is used as M will be described. In this process, a part or the whole of the mark IM for image processing is registered in advance as templates 55a, 55a shown in FIG. Here, the templates 55a and 55b correspond to the first template and the second template according to the present invention, respectively.
FIG. 14 is a diagram illustrating an example of a template used in the correlation method. The template shown in FIG.
It is registered as templates 55a and 55b having a correlation with a part of the Y simultaneous measurement mark MA. here,
The templates 55a and 55b have a mirror image relationship with each other with respect to an axis 56 (predetermined axis) shown in FIG. 14A. Used to measure. Further, reference positions G1 and G2 are set in the respective templates 55a and 55b. Template 5
Since 5a and 55b have a mirror image relationship with respect to the axis 56, the reference positions G1 and G2 also have a mirror image relationship with respect to the axis 56.

The height d2 of the templates 55a and 55b is set to be shorter than the width d1 of the processing range VA shown in FIG. 11, for example, but prevents a plurality of detection results from being obtained at the same position. Therefore, the templates 55a and 55
It is preferable to set the height d2 of b and the width d1 of the processing range VA equal. The main control system 15 is a template 55
a, when a template matching is performed on a certain pattern using 55b, a position where a correlation value between the pattern and the template is equal to or more than a predetermined value and is a maximum is obtained;
Reference position G provided on each template 55a, 55b
1, the midpoint of G2 is detected as the coordinate position of the pattern. The template to be registered is not limited to one line, but may be a template composed of a plurality of lines. Further, FIG. 14A shows a template for detecting the position in the X-axis direction, but a pair of mirror images in the Y-axis direction having the same mirror image relationship as the templates 55a and 55b shown in FIG. A template is provided.
In the template shown in FIG. 14A, 55a and 55b are templates for obtaining the positions of the tops of the edges on both sides of the pattern 57. However, by using the templates 58a and 58b shown in FIG. The position of the bottom of the edge on either side of 57 can be determined. The template 58a and the template 58b have a mirror image relationship with respect to the axis 59 and are the same templates as the templates 55a and 55b, respectively, but are different in the position where the reference position is formed. In FIG. 14C, G11
Is a reference position of the template 58a, and G12 is a reference position of the template 58a.

FIG. 15 is a view showing an image of the image processing mark IM picked up by the image pickup device 23 provided in the aerial image detecting device 20b. As shown in FIG. 15A, when the image processing mark IM formed on the mask M forms an image on the imaging surface of the imaging device 23, the image processing mark IM approximates the templates 55a and 55b within the measurement visual field VF of the imaging device 23. Are detected from the X-axis direction and the Y-axis direction, respectively. When the image processing mark IM is inclined due to rotation or the like, the closer one of the templates 55a and 55b is selected. Then, from the average value of the positions of the respective mark elements constituting the image processing mark IM,
As shown in FIG. 15B, a mark center CP3 of the mark M formed on the image sensor 23 is obtained, and the mark center C3 is determined.
The difference between P3 and the center CP1 of the measurement visual field VF is measured.
At this time, the center CP1 of the measurement visual field VF is a desired position, that is, a design value. Since the position of the plate holder 11 is detected by the laser interferometers 13X and 13Y, the difference between the visual field center CP1 and the mark center CP3 is obtained. Is used to detect the coordinate position of the mark center CP3 in the coordinate system of the plate holder 11.

The configuration of the exposure apparatus having the position measuring device according to the embodiment of the present invention and the plurality of processing algorithms provided in the main control system 15 have been described above. Next, the exposure apparatus according to the embodiment of the present invention will be described. An outline of the operation will be described. First, the outline of the entire operation of the exposure apparatus will be briefly described. The exposure apparatus loads the mask M onto the mask stage MS to determine the baseline amount, and then places the plate P on the plate holder 11 to set the plate. The position information of P is obtained, the relative position between the mask M and the plate P is adjusted, and the device pattern DP formed on the mask M is obtained.
Is transferred to the plate P via the projection optical system PL. Note that the exposure apparatus according to the present embodiment performs exposure processing on the entire plate P using a plurality of masks M, but for simplicity of description, only operations performed on one mask will be described below. I do.

[Measurement of Baseline Amount] FIG. 16 is a flowchart showing a schematic operation of the exposure apparatus according to one embodiment of the present invention when measuring a baseline amount. In the exposure apparatus of this embodiment, the mask M is first placed on the mask holder MS, and the position measurement marks RM drawn on the mask M are observed by the mask observation systems 16a and 16b shown in FIG. Then, the main control system aligns the mask M by adjusting the position of the mask stage MS based on the observation result (Step S10). The reference member FM and the reflection plates RF1 to RF4 provided on the plate stage PS are made to protrude from the plate holder 11. Next, the main control system 15
Moves the plate stage PS, arranges the reference member FM at the center of the projection area of the projection optical system PL, and measures the position information of the plate stage PS in the Z-axis direction by the autofocus mechanism 19 (step S12). Main control system 1
5 is based on the light receiving result output from the light receiving system 19b,
By adjusting the position of the plate stage PS in the Z-axis direction, the reference member FM (corresponding to the upper surface of the plate P) is positioned at a position conjugate with the mask M placed on the mask stage MS.

In the following description, for the sake of simplicity, the focal planes PP of the aerial image detecting devices 20a and 20b are used.
1. The upper surface of the reference member FM when the reference member FM is protruded from the plate holder 11 and the imaging plane FC of the plate alignment sensors 18a to 18d are included in the same plane perpendicular to the Z axis. It is assumed that it is set. If they are different, the shift amount is detected in advance, and the position of the plate stage PS in the Z-axis direction is adjusted so that the shift amount becomes zero as appropriate. For example, the image processing mark I formed on the mask M via the projection optical system PL
When the images of M are detected by the aerial image detecting devices 20a and 20b, the focal plane PP1 of the aerial image detecting devices 20a and 20b and the image forming surface of the projection optical system PL are made to coincide with each other. When detecting the emitted image of the index mark by the aerial image detection devices 20a and 20b, the focal plane PP1 of the aerial image detection devices 20a and 20b and the imaging plane FC of each of the plate alignment sensors 18a to 18d are matched. .

When the above processing is completed, processing for measuring the measurement center of the plate alignment sensors 18a to 18d is performed (step S14). Hereinafter, the details of this processing will be described with reference to FIG. First, the main control system 15 moves the stage PS via the stage drive systems 14X and 14Y, and moves the detection hole AS1 to the plate alignment sensor 1.
8a is arranged in the measurement area. When the arrangement is completed, the plate alignment sensor 18a irradiates the detection hole IL with the detection light IL1 including the image of the index mark. The image of the index mark is picked up by the image sensor 23 included in the aerial image detection device 20a via the detection light AS1, and is output to the main control system 15 as an image signal. FIG. 17 shows an aerial image detecting device 20a.
FIG. 3 is a diagram showing an image of an index mark captured by the image sensor 23 provided in the image forming apparatus.
1 to i4 are respectively attached. When obtaining the position information of the index mark, for example, image processing is performed on an image signal output from the image sensor 23, and first, the processing range V
The edge position (the part where the contrast changes) of each of the partial images i1 to i4 in A (see FIG. 11) is obtained. Here, the edge positions of the partial images i1 to i4 are obtained by the above-described processing of differentiating the image signal, the processing of obtaining the slice level, or the processing by template matching using the templates 55a and 55b shown in FIG.

Next, the midpoint of the position of the partial images i1 and i2 in the X-axis direction is calculated to determine the center position of the index mark image Im in the X-axis direction, and the partial images i3 and i4 in the Y-axis direction are calculated. The center position CP2 of the index mark image is determined by calculating the center of the position and determining the center position of the index mark image Im in the Y-axis direction. Then, the position information of the index mark is measured by calculating the amount of deviation from the center CP1 of the measurement visual field VF.

Here, when the detection hole AS1 is arranged in the measurement area of the plate alignment sensor 18a, the whole image of the index mark is not arranged in the measurement area of the aerial image detection device 20a due to a movement control error of the stage. It is conceivable that an image signal in which a part of the image of the index mark is missing from the element 23 is obtained. FIG. 18 is a diagram illustrating a state in which an image signal in which a part of the index mark is missing is obtained. In the example shown in FIG. 18, the partial image i1, which forms a part of the image of the index mark, is missing. In this state, if the processing described with reference to FIG. 17 is performed, the position of the partial image i1 cannot be obtained, so that the position information of the index mark image cannot be measured, an error occurs, and the processing is restarted. Must be restarted by performing a manual assist process, which causes a decrease in throughput.

In the present embodiment, 3 for forming the index mark is used.
When the position of one partial image is obtained, the position information of the image of the index mark is obtained without interrupting the processing even if the position of one partial image is not obtained. In the example shown in FIG. 18, the position information of the index mark image is measured from the partial images i2, i3, and i4 whose positions can be obtained. In FIG. 18, since the positions of the two partial images in the X-axis direction are obtained from the partial images i3 and i4, positional information of the index mark image Im in the X-axis direction is obtained therefrom. On the other hand, in the Y-axis direction, only one position of the partial image is obtained, but is the position of the partial image indicating the position of the partial image i1 or the partial image i2?
Needs to be determined. This determination is made based on the sign of the value obtained by subtracting the position of the partial image i2 from the center CP1 of the measurement visual field VF. That is, X
Since the positions of the two partial images i3 and i4 in the axial direction are obtained, a part of the partial images i3 and i4 is included in a portion of the measurement area VA having a longitudinal direction in the X-axis direction. In such a case, from the shape of the index mark, the measured position of the partial image indicates the position of the partial image i2 when the sign is negative, and indicates the position of the partial image i1 when the sign is positive. It will be.

When the above processing is completed, one obtained partial image in the Y-axis direction (in the example shown in FIG. 18, partial image i
The position of the partial image (partial image i1 in the example shown in FIG. 18) missing from the position 2), the determination result, and the design value of the index mark is obtained. Since the positions of the two partial images in the Y-axis direction are obtained by this processing, the position information (estimated value) of the index mark image in the Y-axis direction is obtained. Here, since the obtained position information of the image of the index mark is an estimated value, it is considered that there is a problem in its accuracy. Therefore, the main control system 15 moves the stage PS based on the obtained position information (estimated value), and adjusts the relative relationship between the image Im of the index mark and the image sensor 23 provided in the spatial image detecting device 20a.
The position information of the image Im of the index mark is measured again so that the entire image Im of the index mark is formed on the imaging surface of the image sensor 23. As described above, in the present embodiment, the image I of the index mark projected from the plate alignment sensor 18a is provided.
Even if a part of m is missing, the image Im of the index mark is maintained without interrupting the processing and without reducing the measurement accuracy.
Position information can be measured.

After measuring the measurement center of the plate alignment sensor 18a, the main control system 15 moves the stage PS so that the detection hole AS1 is arranged in the measurement area of the plate alignment sensor 18d, and the measurement of the plate alignment sensor 18a is performed. The same processing as the processing for measuring the center is performed to measure the measurement center of the plate alignment sensor 18d. Next, the main control system 15 moves the stage PS so that the detection holes AS2 are
The measurement center of the plate alignment sensor 18c is measured by performing the same processing as described above using the aerial image detection device 20b while being arranged in the measurement area 8c. Finally, the main control system 1
5 moves the stage PS and arranges the detection hole AS2 in the measurement area of the plate alignment sensor 18b to measure the measurement center of the plate alignment sensor 18b. The process of step S14 for measuring the measurement centers of the plate alignment sensors 18a to 18d shown in FIG.
In any case of measuring the measurement center of ~ 18d,
When an image signal in which a part of the index mark is missing is obtained, the processing described with reference to FIG. 18 is performed.

As described above, the plate alignment sensor 18
The aerial image detection devices 20a and 20b
Is compared with the case where all the measurement centers of the plate alignment sensors 18a to 18d are measured by using only one aerial image detecting device, so that the movement amount of the stage PS can be reduced. This is to reduce the size of the PS and shorten the time required for measurement.

When the process of step S14 is completed, a process of measuring the exposure center is performed by the aerial image detecting device 20b (step S16). Here, the exposure center refers to the center of the projected image of the pattern formed on the mask M. In this process, first, the main control system 15 sets the stage drive systems 14X, 1
4Y, the stage PS is moved to the detection hole AS2.
Is the position where the image of the image processing mark IM formed on the mask M is irradiated via the projection optical system PL (this position is determined by the design value of the image processing mark IM, the design magnification of the projection optical system PL, and the like). To the required position). When the movement is completed, exposure light is emitted from the illumination optical system to illuminate the image processing mark IM formed on the mask M. The image of the image processing mark IM is irradiated onto the detection hole AS2 via the projection optical system PL, is picked up by the image sensor 23 provided in the aerial image detection device 20b via the detection hole AS2, and is used as an image signal by the main control system 15. Output to The main control system 15 performs the above-described differential processing or the processing for obtaining the slice level on the image signal output from the image sensor 23 or the processing shown in FIG.
4 is performed using the templates 55a and 55b shown in FIG. 4 to obtain the mark center CP3 of the image of the image processing mark IM shown in FIG. 15, and the mark center CP is determined from the difference between the visual field center CP1 and the mark center CP3. The coordinate position of CP3 is detected.

Here, even when the center of exposure is measured by the aerial image detecting device 20b, the entire image of the image processing mark IM is removed from the aerial image detecting device 2 due to a stage movement control error.
It is conceivable that an image signal in which a part of the image of the index mark is missing from the image sensor 23 is not arranged in the measurement area 0a. FIG. 19 is a diagram illustrating a state in which an image signal in which a part of the image processing mark IM is missing is obtained. In the example shown in FIG. 19, the mark P forming a part of the image processing mark IM
The image of d5 is missing. If any part of the image of the image processing mark IM is missing, the position information of the image of the image processing mark IM cannot be obtained and an error occurs. Must be performed, causing a decrease in throughput.

In order to prevent such a problem, in this embodiment, even if an image signal is obtained in which a part of the image of the mark element forming the image processing mark IM is missing, the processing is interrupted as an error. Instead, the position information of the image processing mark IM is obtained from the remaining mark elements. Hereinafter, the details of this processing will be described. FIG. 20 is a flowchart showing a process for determining the center position of the image processing mark IM performed when a part of the image of the image processing mark IM is missing. First, the image processing mark IM formed on the image sensor 23 included in the aerial image detecting device 20b is imaged with the positional relationship shown in FIG. 19 to obtain an image signal (step S30). Next, image processing is performed on the obtained image signal, and mark elements Pa1 to Pa5, mark element Pb1
To Pb5, mark elements Pc1 to Pc5, and mark elements Pd1 to Pd4, and an approximate center position CP5 of the image processing mark IM is obtained from the detection results (step S32). The innermost mark pair is detected by searching for a pair of marks having an interval that matches the interval between the mark element Pa1 and the mark element Pb1 and the interval between the mark element Pc1 and the mark element Pd1. From the center position CP5 of the mark. At this time, the mark element Pa1 and the mark element Pb1
Is larger than the distance between the mark elements Pa1 and Pa5 and the distance between the mark elements Pb1 and Pb5, and is larger than the distance between the mark elements Pc1 and Pd1.
Are larger than the distance between the mark elements Pc1 and Pc5 and the distance between the mark elements Pd1 and Pd5. In this way, if the marks are arranged so that the interval between the pair of marks to be detected is always different from the interval between other marks, it is possible to find the approximate center position of the mark using that mark pair Becomes

When the approximate center position CP5 of the image processing mark IM is obtained by the above processing, a process of obtaining each mark element pair based on the approximate center position CP5 is performed (step S34). The pair of each mark element is obtained, for example, from the approximate center position CP5 and the relative position between each mark element. For example, in FIG. 19, the two mark elements closest to the approximate center position CP5 in the X-axis direction are a mark element Pc1 and a mark element Pd1. Therefore, the main control system 15 selects and pairs the mark element Pc1 and the mark element Pd1. Similarly, other mark elements arranged in the X-axis direction are selected and paired, and each mark element arranged in the Y-axis direction is selected and paired.

Here, in the example shown in FIG. 19, since the image of the mark element Pd5 is missing, there is no paired mark for the mark element Pc5. In this case, the position of the mark element Pd5 may be estimated from the design value of the image processing mark IM to be paired with the mark element Pc5. Alternatively, the mark element Pc5 may not be taken into account in the process of measuring the position information of the image processing mark IM without finding the pair of the mark elements Pc5. In this case, it is considered that the position measurement accuracy slightly deteriorates because the number of pairs of mark elements decreases. However, from the viewpoint of improving the throughput, some position measurement accuracy can be allowed.

When a pair of mark elements is selected in the above processing, the pair of selected mark element pairs set in the X-axis direction and the Y-axis direction are signal-processed, and the center of the image processing mark IM is processed. A process for obtaining the position CP6 is performed (Step S36). For example, a mark element Pc that forms one of a pair
The midpoint position of 1 and the mark element Pd1 is obtained, the same processing is performed on the other pairs to obtain the midpoint, and the average value of the obtained midpoints is used as the center position CP6 of the image processing mark IM. Processing is performed. Here, in this processing, it is preferable to perform signal processing by weighting the pair of mark elements closer to the approximate center position CP1. This is usually
The position of the image of the image processing mark IM is mostly missing at the periphery of the image processing mark IM, and as described above, the position of the missing mark element is changed to the image processing mark IM.
This is because the accuracy may be lower than the detection accuracy of the mark element closer to the approximate center position CP1. The center position CP
In order to obtain 6, the coordinates of the paired mark elements may all be added, and the coordinates may be divided by the number of the added mark elements.

The above processing is performed, and the position information of the image processing mark IM formed on the mask M is measured by the aerial image detecting device 20b. The mark IM for image processing is a mask M
And a plurality of image processing marks IM.
Are sequentially measured. At this time, the position information of the mask M can be obtained with higher accuracy by increasing the number of the image processing marks IM to be detected. Next, processing for measuring the exposure center is performed by the aerial image detection device 20a (step S18). In the process of step S18, the spatial image detecting device 20 is operated in the same procedure as the process of step S16.
a, a plurality of image processing marks IM formed on the mask M
Is measured. Also in this case, when an image signal in which a part of the image processing mark IM is missing is obtained, FIG.
Is performed.

Then, by performing a statistical calculation such as a least-squares approximation on the measurement result of step S18, correction values such as rotation, shift, magnification and the like required at the time of exposure are obtained (step S20). Center position of mask M in 11 coordinate system (exposure center)
Is required. As described above, in the present embodiment, the detection holes AS1 and AS2 are arranged at the positions where the image of the image processing mark IM formed on the mask M is projected when the exposure center is measured, and then, as in the related art. There is no need to move the stage PS by scanning. Therefore, the time required for measurement can be extremely reduced.

The center of exposure has been measured by each of the spatial image detecting devices 20a and 20b. Next, the spatial image detecting devices 20a and 20b
b is obtained (step S2).
2). In order to obtain the baseline amount, it is sufficient to obtain only the exposure center based on the measurement result of the aerial image detection device 20a. In the present embodiment, since the measurement center of the plate alignment sensors 18a to 18d is measured using the spatial image detecting device 20a and the spatial image detecting device 20b, the relative position between the spatial image detecting device 20a and the spatial image detecting device 20b is measured. In order to accurately determine the relative positional relationship, the plurality of image processing marks IM formed on the mask M are measured by the spatial image detecting devices 20a and 20b, respectively. When the relative positional relationship between the aerial image detecting device 20a and the aerial image detecting device 20b is determined, the main control system 15 determines the relative positional relationship and the measurement center of each of the plate alignment sensors 18a to 18d determined in step S16. The relative positional relationship between the plate alignment sensors 18a to 18d is calculated.

Next, it is determined whether or not the rotation amount and the magnification calculated in step S24 are within a predetermined allowable value (step S24). When it is determined that the value is out of the allowable value (the determination result is “NO”), the main control system 15 outputs a control signal to the driving unit 17 (see FIG. 2) to rotate the mask M. Then, a control signal is output to the lens controller 10 to control the magnification (step S
26). On the other hand, the determination result of step S24 is “YES”
In the case of, the plate alignment sensors 18a to 18d obtained in the processing of step S16 or step S18
The baseline amount is calculated based on the measurement center and the exposure center obtained in the process of step S20 (step S28).

The baseline amount is calculated through the above processing. As described above, the exposure apparatus according to the present embodiment measures the spatial image detecting apparatuses 20a. 20b is used to measure the measurement centers of the plate alignment sensors 18a to 18d and to measure the exposure centers. In the case where the size of the measurement field of view is limited as described above, if an image signal is obtained in which the image of the mark to be measured is partially missing, the position information of the mark cannot be obtained, and an error occurs. Therefore, the processing is interrupted, which is a factor that lowers the throughput. In the present embodiment, the position information can be obtained without interrupting the process regardless of whether the measurement center of the plate alignment sensors 18a to 18d is measured or the center of exposure is measured. Is not reduced. Further, position information can be obtained without significantly lowering accuracy.

[Measurement of Position Information of Plate P] The method of measuring the amount of the baseline has been described above. The operation will be described. FIG. 21 is a schematic diagram showing an arrangement relationship of marks formed on the plate P. As shown in FIG.
And the plate alignment sensors 18a to 18a
Marks AM1 to AM4 having the shapes shown in FIG. 21 are formed at positions corresponding to the respective 18d. Incidentally, FIG.
Areas SA1 to SA4 in the middle indicate areas where the images of the patterns formed on the four masks M are respectively transferred. In FIG. 18, the sizes of the marks AM1 to AM4 are exaggerated. The measurement of the position information of the plate P is performed before the image of the device pattern DP formed on the mask M is transferred to the areas SA1 to SA4. In the present embodiment, it is assumed that the plates P are processed in a lot unit using a plurality of plates P as a unit.

FIG. 22 shows a plate alignment sensor 1.
A formed on plate P using 8a to 18d
It is a flowchart which shows the outline | summary of the operation | movement at the time of measuring the position information of M1-AM4. When one plate P is taken out from one lot, the plate P is pre-aligned and then carried and placed on the plate stage PS.
The image is adjusted to the image planes a to 18d. The reference member FM and the reflection plates RF1 to RF4 are immersed in the plate stage PS before the plate P is placed on the plate plate stage PS. The main control system 15 moves the stage PS to move the mark A formed on the plate P.
M1 to AM4 are plate alignment sensors 18a to 1
It is arranged near the measurement visual field of 8d. At this time, the mark AM1 is applied to the plate alignment sensors 18a to 18d.
ＡＭAM4 are arranged at positions shifted by a predetermined amount (for example, several tens μm) (step S30).

This is to reduce the time required for image processing performed by the main control system 15 on image signals output from each of the plate alignment sensors 18a to 18d. That is, in the search measurement, the position information of the mark is obtained based on the image signal of the image of the mark formed in the measurement areas W11, W12, W21, and W22 shown in FIG. It takes time to perform image processing on all W21 and W22. Therefore, in the X-axis direction, the measurement area W11 or the measurement area W12, Y
In the axial direction, the time required for search measurement is reduced by processing only image signals obtained from the measurement area W21 or the measurement area W22. As described above, when the number of the measurement areas is reduced, the position of the mark with respect to the plate sensors 18a to 18d due to the influence of the magnification (scaling) of the plate P is determined for each plate P (for each mark to be measured). ), A process of setting a measurement area for each mark is required, and the time required for the process is lengthened accordingly. For this reason, the plate alignment sensor 1
By arranging the marks AM1 to AM4 at positions shifted by a predetermined amount with respect to 8a to 18d, the arrangement direction and distance of the marks AM1 to AM4 with respect to the alignment sensors 20a to 18d become constant, so that a measurement area is set for each mark. Required, and as a result, the time required for the measurement process can be reduced.

Incidentally, in the following description and FIG.
The position information of the M1 in the X-axis direction and the Y-axis direction is P1x, P1
y, the position information of the mark AM2 in the X-axis direction and the Y-axis direction is represented by P2x, P2y, respectively, and the position information of the mark AM3 in the X-axis direction and the Y-axis direction is P3x, P
3y, and the position information of the mark AM4 in the X-axis direction and the Y-axis direction are denoted by P4x and P4y, respectively. When the arrangement of the marks AM1 to AM4 is completed, the position information P1x in the X-axis direction and the position information P1y in the y-axis direction of the mark AM1 using the plate alignment sensor 18a and the mark AM using the plate alignment sensor 18b.
A process of searching and measuring the position information P2y in the Y-axis direction 2 is performed (step S32). Here, a case where the position information P1x, P1y, and P2y are measured at the time of the search measurement will be described as an example. In the search measurement, the rotation amount and the shift amount of the plate P may be obtained. The position information in the axial direction and the position information in the two Y-axis directions may be measured, or the position information in the two X-axis directions and the position information in one Y-axis direction may be measured.

FIG. 23 is a diagram showing how the image of the mark AM1 and the image of the index mark are formed on the image pickup surface of the image pickup device 42 provided in the plate alignment sensor 18a during the search measurement. In FIG. 23, FP1 indicates the imaging surface of the imaging element 42 provided in the plate alignment sensor 18a, Im11 indicates the image of the mark AM1 formed on the imaging surface FP1, and Im indicates the image of the index mark formed on the imaging surface FP1. Each is shown. Note that, similarly to FIG. 13, the image of the mark AM2 and the image of the index mark are formed on the imaging surface of the imaging element provided in the plate alignment sensor 18b.
In the example shown in FIG. 23, the mark AM1 is changed to the plate alignment sensor 18 by the processing in step S30 in FIG.
It is arranged to be shifted in the + X axis direction and the + Y axis direction with respect to a. In this case, the plate alignment sensor AM
1, the image sensor 42 provided in AM2 includes a measurement area W11.
And a measurement area W21 are set.

A plate alignment sensor 18a is provided.
The image formed on the imaging plane FP1 of the imaging element 42 is
At 42, it is converted into an image signal and output to the main control system 15.
You. Similarly, the plate alignment sensor 18b has
The image formed on the imaging surface of the image sensor 42 is
The signal is converted into an image signal and output to the main control system 15. Master system
The control system 15 includes plate alignment sensors 18a, 18
b) performs image processing on the image signal output from
Image Im1 formed on each of measurement areas W11 and W12
1, the edge position of Im12 is obtained. The main control system 15
Image sensor 4 included in plate alignment sensor 18a
Measurement areas W11 and W21 set on the second imaging plane FP1
Image sensor provided in plate and plate alignment sensor 18b
The measurement areas W11 and W21 set on the imaging surface 42 are grasped.
Information on the shapes of the marks AM1 and AM2 is also recorded in advance.
I remember. Therefore, mark from the measured edge position
Calculate the center position of AM1 and the center position of mark AM2
You. As described above, the measurement areas W11 and W21 are set and the mark A
When obtaining position information by search measurement of M1 and AM2
Has been described, but the measurement areas W11 and W21 are set.
If the edge position cannot be determined when
The measurement area W11 is switched to the measurement area W12 or the measurement area
The area W21 is switched to the measurement area W22 to change the mark AM1,
Obtain AM2 location information

Here, the extreme distribution of the image signals output from the plate alignment sensors 18a and 18b may change depending on the surface condition of the plate P, illumination conditions, and the like. FIG. 24 is a diagram showing how the mark image observed changes depending on the surface condition of the plate P, illumination conditions, and the like.
In the example shown in FIG. 24A, the image Im1 of the mark AM1
1 is observed as a darker image than the surroundings, and the example shown in FIG. 24B illustrates that the image Im11 of the mark AM1 is observed as an image brighter than the surroundings. FIG.
In the case shown in (a), symbols SX1 and S
An image signal with Y1 is obtained. On the other hand, FIG.
In the case shown in (1), image signals denoted by symbols SX2 and SY2 in the figure are obtained. That is, in FIG. 24A and FIG. 24B, the density relationship of the mark images is reversed.

Now, the image signals SX1, S shown in FIG.
When image processing was performed using a processing algorithm that can obtain the position information of the mark AM1 when Y1 was obtained, the image signals actually obtained were image signals SX2 and SY2 shown in FIG. , The mark A
In some cases, the position information of M1 cannot be obtained. As described above, in the present embodiment, the image processing is performed by first predicting the state of the mark image based on the board information.
When the position information is not obtained contrary to the prediction, the position information shown in FIG.
When the image signal in (b) is SX2 or SY2, image processing is performed by a processing algorithm for obtaining position information of the mark AM1. The main control system 15 includes an alignment sensor 18
The processing algorithm that has performed image processing on the image signal output from the image sensor 42 included in “a” to obtain the position information of the mark AM1 is stored in the storage device 15a as image processing information. As described above, in the present embodiment, the main control system 15 switches a plurality of processing algorithms according to the shading relationship of the mark image included in the image signal.

In this embodiment, for the sake of simplicity, a case will be described in which this image processing information is stored in units of lots including the plate P currently mounted on the plate stage PS. It is considered that the surface condition of the plate P in the same lot or a related lot is almost the same.
Therefore, in the present embodiment, the surface state (the state of the mark image is almost the same) of the other plate P included in the lot and the plate P currently mounted on the plate stage PS.
Is estimated, and the position is measured using the processing algorithm included in the image processing information stored in the storage device 15a, thereby shortening the time required for the position measurement.
It should be noted that the image processing information may be stored with the mark formed on the plate P as a unit with the plate P as a unit.

Returning to FIG. 22, when the processing in step S32 is completed, the position information P1 measured in step S32
A process of calculating the shift amount and the rotation amount of the plate P using x, P1y, and P2y is performed (step S34).
Next, in order to prevent the rotation amount of the plate P from affecting the next measurement processing, it is determined whether the rotation amount of the plate P is within a predetermined allowable range (step S36). .

If the result of the determination in step S36 is "N
In the case of "O", the plate stage PS is rotated so that the rotation amount of the plate P falls within the allowable value in step S38, and the process proceeds to step S40. On the other hand, if the result of the determination in step S36 is "YES", the flow proceeds directly to step S40. In step S40, the position information in the Y-axis direction of the mark AM1 formed on the plate P using the plate alignment sensor 18a, and the position information in the X-axis direction of the mark AM2 formed on the plate P using the plate alignment sensor 18b. Processing for finely measuring the position information in the Y-axis direction is performed.

FIG. 25 shows the image sensor 42 during fine measurement.
FIG. 4 is a diagram showing an example of an image Im11 of a mark AM1 and an image Im of an index mark formed on the imaging surface of FIG. FIG. 25 shows a mark AM1 on the imaging plane FP1 for easy understanding.
3 shows a state where the image Im of the index mark is formed at the center position of the image Im11. Since the image Im of the index mark is formed by shielding the detection light IL1 by the second light shielding portion 37b shown in FIG. 7, the image Im is darker than the surroundings. As shown in FIG. 25, in the fine measurement, the main control system 15 changes the signal intensity in the X-axis direction in the detection area SC1 set in the imaging plane FP1 of the imaging element 42 and changes the Y intensity in the detection area SC2. The edge position of the image formed on the imaging surface FP is detected based on the change in the signal strength in the axial direction, and the mark AM1 relative to the center position of the image Im of the index mark formed on the imaging surface FP is determined from the interval between the edge positions. The position information of the mark AM1 is measured by calculating the shift amounts of the center position of the image Im11 in the X-axis direction and the Y-axis direction. Here, the edge position is detected when measuring the position information of the mark AM1.
Read out the image processing information stored in a. An edge position is detected using a processing algorithm included in the image processing information. Thus, in the present embodiment, the main control system 1
5 gives priority to one of the plurality of processing algorithms based on the image processing information stored in the storage device 15a. Note that the same processing is performed on the mark AM2 to measure the position information. Through the above processing, the main control system 15 obtains the position information P1y, P2x, P2y.

Then, a process for obtaining the shift amount and the rotation amount is performed using the position information P1y, P2x, P2y measured in step S40 (step S42). next,
It is determined whether the rotation amount obtained in step S42 is within an allowable range (step S44). The judgment result is "N
If "O", the stage is rotated to such an extent that the rotation amount of the plate P falls within the allowable range (step S46).
The process returns to step S40. On the other hand, if the determination result of step S44 is "YES", a process of finely measuring the mark AM3 formed on the plate P using the plate alignment sensor 18c is performed (step S44).
48). The main control system 15 also performs this fine measurement.
Performs image processing based on image processing information stored in the storage device 15a.

Through the above processing, the processing for measuring the position information of the marks AM1 to AM3 formed on the plate P using the plate alignment sensors 18a to 18c is completed. As described above, in the present embodiment, information indicating a processing algorithm used when performing image processing is stored in the storage device 15a as image processing information, and when performing image processing, the state of a mark image is predicted and image processing is performed. Since the position information of the mark is obtained, unnecessary processing is not repeatedly performed, so that the time required for image processing can be reduced, and as a result, the throughput can be improved.

The four pieces of position information obtained by the above measurement, that is, the position information P1y of the mark AM1 in the Y-axis direction, the position information P2x of the mark AM2 in the X-axis direction, and the position information P2x of the mark AM2
By using the axial position information P2y and the position information P3x of the mark AM3 in the X-axis direction, a statistical operation process called so-called enhanced global alignment (EGA) measurement is performed, and the array coordinates of the exposure areas SA1 to SA4 are obtained. Is calculated. Next, the main control system 15 determines the exposure area SA of the plate P based on the array coordinates of the exposure areas SA1 to SA4 obtained by the EGA measurement and the base line amount obtained in advance.
1 and the exposure center, the exposure light from the illumination optical system is irradiated onto the mask M, and the image of the pattern formed on the mask M is exposed to the exposure area SA1 of the plate P via the projection optical system PL. Transfer to After that, the mask M is exchanged and the plate stage PS is driven by stepping to expose all the exposure areas SA1 to SA4 set on the plate P. Thus, a series of operations of the exposure processing ends.

The plate P after the exposure processing is carried out of the plate stage PS, a new plate PS is carried in and placed on the plate stage PS, and the above-described search measurement and fine measurement are performed by the plate alignment sensors 18a to 18d. Then, the position information of the plate P is measured. At this time, the plate alignment sensor 18
Image processing is performed on image signals output from a to 18d based on information specifying a processing algorithm included in the image processing information stored in the storage device 15a.

Although the embodiment of the present invention has been described above, the present invention is not limited to the above embodiment, and can be freely modified within the scope of the present invention. For example, in the above embodiment, the step-and-repeat type exposure apparatus has been described as an example, but the present invention is also applicable to a step-and-scan type exposure apparatus. The light source of the illumination optical system of the exposure apparatus according to the present embodiment uses g-ray (436 nm) and i-ray (365 nm) emitted from an ultra-high pressure mercury lamp, but is not limited to this. (248 nm), ArF excimer laser (193n)
m), a laser beam emitted from an F ₂ laser (157 nm), or a charged particle beam such as an X-ray or an electron beam. For example, when an electron beam is used, a thermionic emission type lanthanum hexaborite (LaB ₆ ) or tantalum (Ta) can be used as an electron gun. Further, in the above-described embodiment, the case of manufacturing a liquid crystal display element has been described as an example. An exposure apparatus used to transfer a device pattern onto a semiconductor substrate, an exposure apparatus used to manufacture a thin-film magnetic head to transfer a device pattern onto a ceramic wafer, and an exposure apparatus used to manufacture an imaging device such as a CCD. The present invention can also be applied to

The index marks formed on the index plate provided in the position measuring device of the present invention are not limited to those shown in FIG. The marks AM1 to AM1 formed on the plate P
AM4 is not limited to the shape shown in FIG. Index marks and marks AM1 to AM formed on plate P
The shape of M4 can be appropriately designed according to processing such as image processing. For example, in the above embodiment, the relationship between the mark image Im11 and the index mark image Im when formed on the imaging surface of the image sensor is such that the index mark image Im is the mark image as shown in FIG. Although the relationship is surrounded by Im11, the relationship may be designed so that the mark image Im11 is surrounded by the index mark image Im. When the position information of the mark AM formed on the plate P is measured by the plate alignment sensors AM1 to AM4, the surface state of the plate P (for example, the shape of the pattern formed on the surface or the photosensitive agent applied to the surface). The combination of the dichroic filters 32 shown in FIG.
Is preferably changed.

In the above embodiment, when the spatial image detecting devices 20a and 20b measure the position information of the image of the image processing mark IM or the index mark image, the image signal with a part of the image missing is measured. Even when it is obtained, the case where the position information of the mark is measured has been described as an example. However, when the position information of the marks AM1 to AM4 formed on the plate P is measured by the plate alignment sensors 18a to 18d. Similarly, the marks AM1 to AM4 can be obtained by performing the same processing. Further, in the above embodiment, the marks AM1 to AM1
The position information of AM4 is transmitted to the plate alignment sensor 18a.
In the case where the position information is measured by changing the processing algorithm and performing image processing even when the state of the image obtained by the surface state of the plate P or the illumination condition changes when measuring at ~ 18d. As described above, when the image Im of the index mark is measured, it is possible to cope with the case where the shading of the image changes according to the surface condition of the plate P or the illumination condition by changing the processing algorithm.

Next, an embodiment of a method for manufacturing a micro device using an exposure apparatus according to an embodiment of the present invention in a lithography process will be described. FIG. 26 shows a micro device (a semiconductor chip such as an IC or an LSI, a liquid crystal panel,
FIG. 4 is a diagram showing a flowchart of a manufacturing example of a CCD, a thin-film magnetic head, a micromachine, and the like). As shown in FIG. 26, first, in step S50 (design step),
The function / performance design of the micro device (for example, the circuit design of a semiconductor device) is performed, and the pattern design for realizing the function is performed. Subsequently, in step S51 (mask manufacturing step), a mask (reticle) on which the designed circuit pattern is formed is manufactured. On the other hand, in step S52 (wafer manufacturing step), a wafer is manufactured using a material such as silicon.

Next, in step S53 (wafer processing step), using the mask and wafer prepared in steps S50 to S52, an actual circuit or the like is formed on the wafer by lithography or the like, as described later. . Next, in step S54 (device assembly step), device assembly is performed using the wafer processed in step S53. Step S54 includes processes such as a dicing process, a bonding process, and a packaging process (chip encapsulation) as necessary.
Finally, in step S55 (inspection step), inspections such as an operation confirmation test and a durability test of the microdevice manufactured in step S54 are performed. After these steps, the microdevice is completed and shipped.

FIG. 27 is a diagram showing an example of a detailed flow of step S53 in FIG. 26 in the case of a semiconductor device. In FIG. 27, in step S61 (oxidation step), the surface of the wafer is oxidized. In step S62 (CVD step), an insulating film is formed on the wafer surface. Step S63 (electrode forming step)
In, electrodes are formed on a wafer by vapor deposition. In step S64 (ion implantation step), ions are implanted into the wafer. Each of the above steps S61 to S64 constitutes a pre-processing step in each stage of the wafer processing, and is selected and executed according to a necessary process in each stage.

In each stage of the wafer process, when the above-mentioned pre-processing step is completed, the post-processing step is executed as follows. In this post-processing step, first, in step S
At 65 (resist forming step), a photosensitive agent is applied to the wafer. Subsequently, in step S66 (exposure step), the circuit pattern of the mask is transferred to the wafer by the lithography system (exposure apparatus) and the exposure method described above. Next, in Step S67 (developing step), the exposed wafer is developed, and Step S67 is performed.
In step 68 (etching step), the exposed member other than the portion where the resist remains is removed by etching. Then, in step S69 (resist removing step), unnecessary resist after etching is removed. By repeating these pre-processing and post-processing steps, multiple circuit patterns are formed on the wafer.

If the microdevice manufacturing method of the present embodiment described above is used, the above-described exposure apparatus and the above-described exposure method are used in the exposure step (step S66), and the resolution is improved by the illumination light in the vacuum ultraviolet region. In addition, since the exposure amount can be controlled with high precision, a highly integrated device having a minimum line width of about 0.1 μm can be produced with a high yield.

In addition to a micro device such as a semiconductor device, a reticle or a mask used in an optical exposure apparatus, an EUV exposure apparatus, an X-ray exposure apparatus, an electron beam exposure apparatus, or the like is manufactured using a mother reticle. The present invention is also applicable to an exposure apparatus that transfers a circuit pattern to a glass substrate, a silicon wafer, or the like. Here, in an exposure apparatus using DUV (deep ultraviolet) or VUV (vacuum ultraviolet) light, a transmission reticle is generally used, and as a reticle substrate, quartz glass, fluorine-doped quartz glass, fluorite, Magnesium fluoride, quartz, or the like is used. In a proximity type X-ray exposure apparatus, electron beam exposure apparatus, or the like, a transmission mask (stencil mask, membrane mask) is used, and a silicon wafer or the like is used as a mask substrate. Such an exposure apparatus is described in WO99
No./34255, WO99 / 50712, WO99 /
66370, JP-A-11-194479, JP-A-2000-12453
And JP-A-2000-29202.

[0092]

As described above, according to the present invention, when image processing is performed on an image signal output from an image sensor, image processing is performed by predicting the state of a mark image based on substrate information. Thus, position information of the mark image provided on the substrate is obtained. Therefore, image processing is optimized for a lot-by-lot substrate, and the situation that mark position information is not required can be eliminated or minimized, so that the mark position can be measured in a short time. As a result, there is an effect that the throughput can be improved. Also, according to the present invention, template matching is performed on an image signal using a first template and a second template that are mirror images of each other with respect to a predetermined axis, and mark position information is obtained from each reference position. Since it is required, there is an effect that the position of the mark can be measured accurately and in a short time. According to the present invention, first, an approximate center position of an alignment mark is obtained, a pair of symmetrically arranged mark elements is selected, and signal processing is performed on the selected pair of mark elements to obtain position information of the alignment mark. For example, even if an image signal in which a part of the alignment mark is missing is obtained,
There is an effect that the position information of the alignment mark can be obtained from the remaining mark elements constituting the alignment mark without greatly deteriorating the accuracy. Also,
According to the present invention, since the processing algorithm used at the time of position detection by imaging an alignment mark is stored, when processing another substrate in the same lot or a related lot, the processing algorithm is used. When the processing is performed using the method, the situation that the position information of the alignment mark cannot be obtained can be eliminated or extremely reduced, so that there is an effect that the position information of the alignment mark can be obtained in a short time.

[Brief description of the drawings]

FIG. 1 is a perspective view showing a schematic configuration of an exposure apparatus including a position measuring device according to an embodiment of the present invention.

FIG. 2 is a front view showing a schematic configuration of an exposure apparatus including a position measuring device according to an embodiment of the present invention.

FIG. 3 is a diagram showing an arrangement of members provided on a plate holder 11 used when measuring various types of position information.

FIG. 4 shows an image processing mark IM formed on a mask M.
It is a figure showing an example of shape of.

FIG. 5 shows plate alignment sensors 18a to 18d.
FIG. 2 is a diagram showing a configuration of an optical system of FIG.

FIG. 6 is a diagram illustrating an example of the shape of an index mark 37 formed on an index plate 36;

FIG. 7 is an enlarged view of a second light shielding portion 37b.

FIG. 8 is a diagram showing a measurement area for search measurement and a measurement area for fine measurement.

FIG. 9 is a view showing the shape of a mark AM formed on a plate P.

FIG. 10 is a cross-sectional view illustrating a configuration of the aerial image detection devices 20a and 20b.

FIG. 11 is a diagram for explaining a process of measuring position information of an IM of an image processing mark using the aerial image detection devices 20a and 20b.

FIG. 12 is a diagram for explaining a differential process performed on an image signal.

FIG. 13 is a diagram illustrating a slice process performed on an image signal.

FIG. 14 is a diagram illustrating an example of a template used in the correlation method.

FIG. 15 illustrates an image sensor 2 included in the aerial image detection device 20b.
FIG. 3 is a diagram illustrating an image of an image processing mark IM captured in 3.

FIG. 16 is a flowchart illustrating a schematic operation of the exposure apparatus according to the embodiment of the present invention when measuring a baseline amount.

FIG. 17 shows an image sensor 2 included in the aerial image detection device 20a.
FIG. 3 is a diagram showing an image of an index mark captured in 3.

FIG. 18 is a diagram illustrating a state in which an image signal in which a part of an index mark is missing is obtained.

FIG. 19 is a diagram illustrating a state in which an image signal in which a part of the image processing mark IM is missing is obtained.

FIG. 20 is a flowchart illustrating a process of obtaining a center position of the image processing mark IM performed when a part of the image of the image processing mark IM is missing.

FIG. 21 is a schematic diagram showing an arrangement relationship of marks formed on a plate P.

FIG. 22 shows plate alignment sensors 18a to 18
marks AM1 to AM formed on plate P using
4 is a flowchart illustrating an outline of an operation when measuring position information of No. 4;

FIG. 23 is a diagram illustrating a state in which an image of a mark AM1 and an image of an index mark are formed on an imaging surface of an imaging element provided in a plate alignment sensor during a search measurement.

FIG. 24 is a diagram showing how a mark image observed changes depending on the surface condition of the plate P, illumination conditions, and the like.

FIG. 25 is a diagram illustrating an example of an image Im11 of a mark AM1 and an image Im of an index mark formed on the imaging surface of the imaging element during fine measurement.

FIG. 26 is a view illustrating a flowchart illustrating an example of a manufacturing process of a micro device.

FIG. 27 is a diagram showing an example of a detailed flow of step S53 in FIG. 26 in the case of a semiconductor device.

[Explanation of symbols]

15 Main control system (processing means) 15a Storage device (storage means,
Image processing unit) 34 Optical fiber (optical system) 35 Condenser lens (optical system) 36 Indicator plate (optical system) 38 Relay lens (optical system) 39 Half mirror (optical system) 40 Objective lens (optical system) 41 Second objective Lens (optical system) 42 Image sensor 55a, 58a Template (first template) 55b, 58b Template (second template) 56, 59 Axis (predetermined axis) AM1 to AM4 Mark (alignment mark) G1, G2, G11 , G12 Reference position IM Image processing mark (alignment mark) P plate (substrate) Pa1 to Pa5 mark element Pb1 to Pb5 mark element Pc1 to Pc5 mark element Pd1 to Pd5 mark element

Continued on the front page F term (reference) 2F065 AA03 AA04 AA06 AA12 AA17 AA20 BB27 CC00 CC20 CC25 DD06 FF04 FF10 FF41 FF51 FF61 GG02 GG04 GG24 HH12 HH13 JJ01 JJ03 JJ05 JJ08 JJ09 Q12 LL12 Q12 LL13 QQ38 QQ41 QQ42 SS02 SS13 TT02 5B057 AA03 CA08 CA12 CA16 CC02 CH18 DA07 DC16 DC33 5F031 CA05 CA07 HA57 JA02 JA04 JA06 JA12 JA17 JA28 JA29 JA38 KA06 MA27 5F046 DD03 FA10 FC04

Claims (12)

[Claims] An optical system for projecting a mark provided for each substrate on a substrate in a lot unit, an image sensor for outputting an image signal obtained by photoelectrically converting an optical image projected via the optical system, Processing means for performing image processing on an image signal output from the image sensor and obtaining position information of a mark image included in the optical image, wherein the processing means A position measuring device for predicting a state of a mark image of the image signal based on information and performing image processing to obtain position information of the mark. 2. The image processing apparatus according to claim 1, wherein the processing unit has a plurality of image processing algorithms for detecting an edge portion of the mark, and switches the plurality of image processing algorithms in accordance with a density relationship of a mark image of the image signal. The position measuring device according to claim 1, wherein: 3. A storage unit for storing image processing information in image processing of the image signal and substrate information on a lot-by-lot basis, wherein the processing unit is configured to store the image processing information based on the image processing information stored in the storage unit. 3. The position measuring device according to claim 2, wherein one of said plurality of image processing algorithms is prioritized. 4. The position measuring apparatus according to claim 1, wherein said storage means stores image processing information for each substrate of said lot for performing position measurement. 5. The image signal according to claim 1, wherein the image signal includes a plurality of mark images, and the storage unit stores image processing information for each of the mark images. Item 2. The position measuring device according to item 1. 6. The optical system projects an index serving as a measurement reference near a mark provided on the substrate, and displays an image of the projected index and an image of the mark on a detection surface of the image sensor. The position measuring device according to claim 1, wherein the position is measured. 7. An optical system for projecting a mark provided on a substrate, an image sensor for outputting an image signal obtained by photoelectrically converting an optical image projected via the optical system, and an image output from the image sensor. A position measuring apparatus comprising: a signal processing unit that performs image processing of a signal and obtains position information of a mark image included in the optical image, wherein a reference position is provided at a predetermined position in a correlation with a part of the mark image. A first template, a second template formed in a mirror image relationship with the first template with respect to a predetermined axis and the reference position, and a first template with respect to each symmetric position of the mark image. An image processing unit that performs pattern matching between the template and the second template and obtains a mark position from a reference position provided in each template. Position measurement device. 8. An alignment method for detecting a position of an alignment mark provided on a substrate, comprising: a first step of imaging the alignment mark formed by symmetrically arranging a plurality of mark elements with respect to a predetermined axis. A second step of obtaining an approximate center position of the imaged alignment mark; and a third step of selecting a pair of symmetrically arranged mark elements based on the center position obtained in the second step; Obtaining a center position of the alignment mark by performing signal processing on the selected pair of mark elements. 9. The signal processing according to claim 4, wherein the fourth step performs signal processing on the pair of mark elements arranged symmetrically with a weight on a pair of mark elements closer to the center of the alignment mark. The alignment method according to claim 8, wherein the alignment is performed. 10. The alignment mark, wherein a plurality of pairs of mark elements are arranged symmetrically with respect to a direction orthogonal to the predetermined axis, and the center positions in the respective directions are substantially coincident with each other. 10. The alignment method according to claim 8, wherein a center position of the alignment mark is obtained by performing the third step and the fourth step with respect to a direction. 11. An alignment method for detecting a position of an alignment mark provided for each substrate with respect to a substrate in a lot unit, wherein an image of the alignment mark of the substrate is taken, and the position of the imaged alignment mark is detected. An alignment method characterized by storing a used processing algorithm. 12. The method according to claim 11, wherein, in the alignment of the substrates in the lot unit, position detection is performed using a processing algorithm of a successful position detection in a previous substrate of the same lot or a related lot. Alignment method.