JPH10172900A - Exposure apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To keep high throughput and also maintain precise alignment accuracy even when foreign particles, etc., adhere to a surface of a photosensitive substrate. SOLUTION: An alignment sensor with a detection area equal to an alignment mark on a wafer W and a focus sensor for detecting defocus in the area by illuminating the area are built in a sensor unit 16. The sensor unit 16 measures the out-of-forcus for detection of the alignment. In a main control device 24, a shot arrangement of the wafer W is computed by EGA(Enhanced Global Alignment), so called, using only the measurement result of mark positions with the defocus within the prescribed range. The exposure position of a shot area is determined to expose a reticle pattern.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure apparatus,
More specifically, the present invention relates to an exposure apparatus used in a photolithography process for manufacturing, for example, a semiconductor element or a liquid crystal display element.

[0002]

2. Description of the Related Art When a semiconductor element or a liquid crystal display element is manufactured by a photolithography process, an image of a pattern of a photomask or a reticle (hereinafter, collectively referred to as a "reticle") is exposed to a photosensitive material through a projection optical system. There is used a projection exposure apparatus that projects onto each shot area on a substrate such as a wafer or a glass plate (hereinafter, collectively referred to as a “wafer”) on which is coated. In recent years, as a projection exposure apparatus of this type, a wafer is mounted on a two-dimensionally movable stage, and the wafer is moved (stepped) by this stage, and an image of a reticle pattern is shot on each wafer. 2. Description of the Related Art A so-called step-and-repeat type exposure apparatus that repeats an operation of sequentially exposing an area, particularly, a reduction projection type exposure apparatus (stepper) is relatively frequently used.

[0003] For example, semiconductor elements are formed by superposing and exposing a multi-layer circuit pattern on a wafer.
Then, when projecting and exposing the circuit pattern of the second and subsequent layers on the wafer, the alignment between the already formed circuit pattern on the wafer and the image of the reticle pattern, that is, the alignment between the wafer and the reticle (alignment) ) Must be performed accurately. For this alignment, a mark for position detection (alignment mark) is formed on the wafer together with the existing circuit pattern, and the position of the circuit pattern can be accurately recognized by detecting the position of the mark with an alignment sensor. Can be.

[0004] The position of a stage on which a wafer is mounted (hereinafter, referred to as a "wafer stage") is precisely measured using a laser interferometer, and the position of the stage during alignment is accurately measured. The overlay exposure is performed by adjusting the stage position so that each shot area is correctly aligned with the exposure position.

In a conventional exposure apparatus, focus measurement (measurement of the position of the projection optical system in the optical axis direction on the wafer surface) is performed in a range of about the exposure field size, and the wafer surface is positioned at an average focus position in that range. Was equipped with an autofocus device for adjusting the position in the optical axis direction. The flatness of a substrate to be exposed, such as a wafer, can be measured by using the focus measurement result obtained by the autofocus device.

[0006]

Incidentally, the degree of roughness of the wafer surface changes due to the exposure and subsequent processes, and the alignment mark (wafer mark) and the surrounding base material are changed depending on the layer on the wafer. In order to accurately detect the position of the alignment mark and perform accurate alignment, the detection area of the alignment sensor needs to match (focus) the focal plane of the objective lens. there were.

However, in the above-described conventional auto-force apparatus in which focus measurement is performed within a range of about the exposure field size and focus is adjusted to an average focus position of the exposure field, a small area about the detection area by the alignment sensor is used. It has been difficult to measure the true defocus amount and accurately adjust the focus of this minute area.

In a step-and-repeat type (or step-and-scan type) exposure apparatus, a projection image of a reticle pattern is aligned with a chip pattern in a shot area on a wafer. As disclosed in, for example, Kaikai 61-44429,
Specific multiple shots on the wafer (sample shots)
The position of the alignment mark attached to the wafer is measured using an alignment sensor, and all shot array coordinates on the wafer are obtained by statistical processing using the least squares method based on the measured value and the design value of each shot array. In many cases, enhanced global alignment (hereinafter, referred to as “EGA”) is employed.

However, as described above, it is difficult for the conventional autofocus apparatus to measure the true defocus amount in a minute area of about the detection area by the alignment sensor and accurately adjust the focus of the minute area. Therefore, the alignment mark area where foreign matter such as dust adheres to the surface thereof or the shot area where a marked step is generated between each mark (in the case of multi-mark) in the alignment mark area due to the rough surface of the wafer are described above. There is a possibility that inconvenience such as selection as a sample shot may occur. In such a case, a position measurement error of the alignment mark occurs, and as a result, an error occurs in shot array coordinates obtained by EGA.

The present invention has been made under such circumstances, and an object of the invention described in claims 1 to 3 is to maintain a high throughput, and to superimpose even a foreign substance or the like adheres to the surface of a photosensitive substrate. An object of the present invention is to provide an exposure apparatus capable of maintaining alignment accuracy with high accuracy.

[0011]

According to the first aspect of the present invention, an image of a pattern formed on a mask (R) is projected onto a projection optical system (W) while a photosensitive substrate (W) is sequentially positioned at a predetermined exposure position. An exposure apparatus for sequentially exposing a plurality of shot areas (SA) on the photosensitive substrate (W) via the PL)
A detection area (86B, 8B) which is approximately the same as the alignment mark area provided to each shot area on the photosensitive substrate (W)
6C) and a mark position detecting means (30-66, 68X, 6) for detecting the position of the alignment mark (90X, 90Y) located in the detection area (86B, 86C).
8Y, 18); irradiating the detection area (86B, 86C) with a light beam and receiving the reflected light to detect a shift of the area with respect to the in-focus surface and based on the shift amount, Focusing means (32, 38 to 52, 7) for adjusting the position of the substrate (W) in the optical axis direction (Z direction)
0 to 78, 14, 22, 24); detection of the position of the alignment mark detected in the in-focus state by the mark detection means, wherein the shift amount detected by the focus means is within a predetermined range. Calculating means (24) for calculating an array of said plurality of shot areas on said photosensitive substrate by statistical processing using a least square method based on said detection results and array data of shot areas in a design, using only the result; Having.

According to this, prior to exposure, each time the alignment mark area provided to each shot area on the photosensitive substrate is positioned in the detection area of the mark position detection means, the position of the alignment mark is determined. Are sequentially detected by the mark position detecting means. Prior to each detection, the focusing unit irradiates the detection area with a light beam and receives the reflected light to detect a shift of the area with respect to the focus plane, and based on the amount of the shift, Adjust the position of the substrate in the optical axis direction. In other words, the position of the photosensitive substrate surface is adjusted by the focusing means, and the position of the alignment mark attached to each shot area in the focused state is detected.

The calculating means uses only the detection result of the position of the alignment mark detected in the in-focus state by the mark detecting means, wherein the shift amount detected by the focusing means is within a predetermined range, The arrangement of a plurality of shot areas on the photosensitive substrate is calculated by statistical processing using the least squares method based on the detection result and the designed arrangement data of the shot areas.

For this reason, the detection result of the position of the alignment mark in which the foreign matter such as dust adheres and the shift amount detected by the focusing means is out of the predetermined range is excluded from the basic data of the shot array calculation. Since only the detection result of the position of the alignment mark in which the amount of displacement detected by the focusing means is within a predetermined range without the attachment of foreign matter or the like becomes the basic data of the calculation of the shot array, it was calculated by the calculating means. The arrangement of the shot areas is accurate without any error due to the attachment of foreign matter and the like.

Then, based on the calculated array data of the shot areas, the photosensitive substrate is sequentially positioned at a predetermined exposure position, and an image of a pattern formed on the mask is transferred to a plurality of images on the photosensitive substrate via a projection optical system. Are sequentially exposed.

Therefore, foreign matter and the like can be kept on the surface of the photosensitive substrate while maintaining a higher throughput than in the case of the so-called die-by-die method in which the position of the alignment mark is detected every time each shot area is exposed. Even when they adhere, it is possible to maintain the overlay accuracy with high accuracy.

In this case, the positioning marks of a plurality of specific shots are determined in advance as detection targets, and position detection is sequentially performed on the positioning marks of the specific plurality of shots in an in-focus state. The array of shot areas may be calculated by statistical processing using the least squares method using only the detection results of the alignment marks within a predetermined range. However, the present invention is not limited to this. Is determined as a detection target, and the amount of defocus detected by the focusing unit with respect to the detection target alignment mark (or the alignment mark of an arbitrary shot area) is within a predetermined range. Only in this case, the position detection is performed sequentially in the focused state, and when the position detection result reaches a predetermined number, the position detection result is used. The sequence of-shot area may be calculated by statistical processing using the method of least squares.

In the above case, the processing of (movement of photosensitive substrate → focusing (focusing) → position detection) is performed without determining whether or not the defocus amount is within a predetermined range at the time of position detection. The position detection sequence of the alignment mark can be configured by repeating the above, and the control content is relatively simple, but it is necessary to set the number of alignment marks to be detected somewhat larger, and Since the measurement is performed for all the positioning marks determined as the detection targets, it takes a long time to perform the measurement. On the other hand, in the above case, it is necessary to determine whether or not the defocus amount is within a predetermined range when detecting the position of each alignment mark. Since the process of judging whether or not the shift amount is within a predetermined range → focus position adjustment (focusing) → position detection is repeated, the control content is somewhat complicated as compared with
Since the measurement can be completed when the number of position detection results of the alignment marks reaches a predetermined number, the measurement time is shorter than in most cases, and an improvement in throughput can be expected.

In this case, as long as the mark position detecting means and the focusing means can match the detection area, the objective lenses of the respective detection optical systems may be different, but as in the second aspect of the present invention. The focusing means may irradiate the light beam and receive the reflected light via the same objective optical system (48) as the mark position detecting means. In this case, even if the imaging characteristic of the objective optical system changes, it is not affected by the change.
It is possible to accurately detect the defocus of the alignment mark area.

In addition, the detection optical system of the focusing means can detect the displacement of the area with respect to the focusing plane by irradiating the detection area of the mark position detection means with a light beam and receiving the reflected light. The focusing means may be of any configuration, but as in the invention according to claim 3,
An optical member (76) for breaking the telecentricity of the reflected light when receiving the reflected light may be provided. In this case, the focus shift (displacement in the height direction) of the alignment mark area is detected by converting the focus shift (displacement in the height direction) into a lateral shift of the re-imaged image with a simple configuration in which the optical member is arranged on a predetermined surface. It becomes possible. Here, as an optical member that breaks the telecentricity of the reflected light, a light-shielding plate disposed on a pupil surface for pupil restriction, a so-called pupil division filter, or the like can be used.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to FIG.
7 will be described with reference to FIG.

FIG. 1 shows an exposure apparatus 10 according to one embodiment.
Is schematically shown. The exposure apparatus 10 is a step-and-repeat type reduction projection exposure apparatus (so-called stepper).

The exposure apparatus 10 includes an illumination system IOP for illuminating a reticle R as a mask with exposure illumination light, a reticle stage RST for holding the reticle R, and an image of a pattern formed on the reticle R as a photosensitive substrate. Optical system PL for projecting onto the wafer W, and the optical system PL
An XY stage 12 that can be moved two-dimensionally in a reference plane (XY plane in FIG. 1) below the XY stage 12, and a minute range (in the Z-axis direction orthogonal to the XY plane while holding the wafer W mounted on the XY stage 12) (For example, within a range of 100 μm) and a sample stage 14 that can rotate within a minute angle range, and an alignment mark (wafer mark) as a position detection mark formed on the wafer W. A sensor unit 16 in which an axis alignment sensor and a focus sensor are integrated, and an alignment control unit 18 for calculating a focus position and an alignment mark position based on an output signal from the sensor unit 16 as described later. A laser interferometer 20 for measuring the position of the sample table 14 in the XY plane, the XY stage 12 and the sample table 14
And a main controller 24 composed of a microcomputer (or minicomputer) for controlling the entire apparatus as a whole.

The illumination system IOP includes a light source (such as a mercury lamp or an excimer laser), a fly-eye lens, a relay lens,
And an illumination optical system including a condenser lens and the like. This illumination system IOP includes illumination light I for exposure from a light source.
L illuminates the pattern CCr on the lower surface (pattern forming surface) of the reticle R with a uniform illuminance distribution. Here, as the exposure illumination light IL, a bright line such as an i-line of a mercury lamp, or an excimer laser beam such as KrF or ArF is used.

The reticle R is fixed on the reticle stage RST by, for example, vacuum suction. The reticle stage RST is driven by a drive system (not shown) in the X direction (the horizontal direction in FIG. 1) and the Y direction (in FIG. 1). It can be minutely driven in a direction perpendicular to the paper surface) and in a θ direction (a rotation direction in the XY plane). Thus, reticle stage RST positions reticle R in a state where the center (reticle center) of pattern CCr of reticle R substantially coincides with the optical axis of projection optical system PL (reticle alignment).
I can do it. FIG. 1 shows a state in which this reticle alignment has been performed.

The projection optical system PL has its optical axis AX set in the Z-axis direction orthogonal to the moving surface of the reticle stage RST. In this case, both sides are telecentric and a predetermined reduction magnification β (β is, for example, １ ／) is obtained. Are used. Therefore, as described later, the reticle pattern CC
When the reticle R is illuminated with uniform illuminance by the illumination light IL in a state where alignment between the r and the shot area on the wafer W has been performed (alignment), the pattern CCr on the pattern formation surface is projected by the projection optical system PL. The image is reduced at the reduction magnification β and projected onto the wafer W coated with the photoresist, and a reduced image of the pattern is formed in each shot area on the wafer W.

The XY stage 12 is actually a Y stage that moves on a base (not shown) in the Y direction (a direction orthogonal to the plane of FIG. 1), and a XY stage 12 that moves on the Y stage (the horizontal direction of the plane of FIG. 1). ), Which are shown in FIG. 1 as an XY stage 12 as a representative.

A sample stage 14 is mounted on the XY stage 12, and is minutely driven by a drive system 22 in the Z-axis direction and the θ-direction (rotation direction around the Z-axis).
A wafer W is suction-fixed onto the sample table 14 via a wafer holder (not shown).

A movable mirror 26 is provided on the upper surface of the sample table 14, and the position of the sample table 14 in the XY plane is projected by projecting a laser beam on the movable mirror 26 and receiving the reflected light. The laser interferometer 20 to be measured is a movable mirror 26
Are provided so as to face the reflection surfaces of the first and second mirrors. In addition, the moving mirror
In fact, an X moving mirror having a reflecting surface orthogonal to the X axis,
A Y moving mirror having a reflecting surface orthogonal to the axis is provided, and correspondingly, a laser interferometer is provided with an X laser interferometer for measuring the X direction position and a Y laser interferometer for measuring the Y direction position. However, in FIG. 1, these are typically shown as a moving mirror 26 and a laser interferometer 20. Therefore, in the following description, it is assumed that the XY coordinates of the sample stage 14 are measured by the laser interferometer 20.

A reference plate FP is fixed on the sample stage 14 so that the surface is at the same height as the surface of the wafer W. The surface of the reference plate FP is used for baseline measurement (measurement of an interval between an optical axis AX of the projection optical system PL and a detection center of an alignment sensor (which will be described later) configuring the sensor unit 16) and the like. Various reference marks including the reference mark are formed.

The measured value of the laser interferometer 20 is
The main controller 24 positions the sample table 14 in the XY two-dimensional directions via the drive system 22 and the XY stage 12 while monitoring the measurement values of the laser interferometer 20. In addition, the main controller 24 also outputs the output of a not-shown oblique incident light type focus sensor (for performing focus measurement in the range of about the exposure field size and performing focus adjustment to an average focus position of the exposure field). The main controller 24 drives the sample stage 14 in the Z direction (focus direction) via the drive system 22 based on the output of the oblique incident light type focus sensor during exposure. That is, the sample stage 14
The wafer W is positioned in the three axial directions of X, Y, and Z via.

In the present embodiment, the main controller 24
Controls the Z drive amount of the sample stage 14 based on the focal position information from the alignment control unit 18 described later to perform focusing on the objective lens of the alignment sensor described later, which constitutes the sensor unit 16. Further, main controller 24 controls the operation of drive system 22 based on the position information of the wafer mark (alignment mark) detected by alignment control unit 18 to control XY stage 1.
2 is stepped to align the center of each shot area of the wafer W with the optical axis AX of the projection optical system PL, and expose the pattern of the reticle R. The operation at the time of exposure will be described later in detail.

The sensor unit 16 includes a projection optical system P
L, and the sensor unit 16
Is configured by integrating an off-axis type alignment sensor and a focus sensor. And
An external halogen lamp 2 is provided in the sensor unit 16.
Illumination light is guided from the optical fiber 8 via the optical fiber 30, various light beams from the sensor unit 16 are irradiated on the wafer W via the prism mirror 32, and the reflected light from the wafer W is reflected via the prism mirror 32. Sensor unit 16
Has been returned to. Various detection signals from the sensor unit 16 are supplied to the alignment control unit 18.

Next, an alignment sensor and a focus sensor constituting the sensor unit 16 will be described in detail with reference to FIG. FIG. 2 shows the internal structure of a sensor unit 16 including an alignment sensor and a focus sensor.

As the alignment sensor, here, F
An IA (Field Image Alignment) type alignment sensor is used. The alignment sensor includes an optical fiber 30, a condenser lens 34, a field stop plate 3
6, dichroic mirror 38, lens system 40, beam splitter 42, prism mirrors 44 and 46, objective lens 48 as objective optical system, lens system 50, dichroic mirror 52, mirror 56, index plate 58, imaging lens 60 , A mirror 62, a relay lens 64, a beam splitter 66, a two-dimensional image sensor 68 X for the X-axis and a two-dimensional image sensor 68 for the Y-axis, such as a two-dimensional CCD.
Y and the like.

Here, each component of the alignment sensor will be described together with its operation.

From the optical fiber 30, broadband illumination light L which is insensitive to the photoresist on the wafer W
The illumination light L3 is emitted from the condenser lens 3
4 illuminates the field stop plate 36 with uniform illuminance. Illumination light restricted by the field stop plate 36 is reflected by the dichroic mirror 38 and passes through the lens system 40 to enter the beam splitter 42. The illumination light split by the reflection at the beam splitter 42 is sequentially reflected by the prism mirror 44 and the prism mirror 46,
8, and a predetermined area on the wafer W is illuminated via the prism mirror 32. Here, in FIG. 2, it is assumed that a wafer mark exists in this predetermined area.

In the illumination light transmission path for the wafer, the field stop plate 36 is conjugate to the wafer W with respect to the combined system of the lens system 40 and the objective lens 48 (image formation relationship). The illumination light L3 that has passed through the field stop plate 36 is the illumination light of the alignment system (FIA system), and the illumination area for the wafer W by the FIA system is uniquely defined by the shape and size of the opening formed in the field stop plate 36. Is determined.

Then, of the illumination light from the optical fiber 30, light reflected by the wafer W (regular reflected light, scattered light, etc.)
Returns to the beam splitter 42 via the prism mirror 32, the objective lens 48, and the prism mirrors 46 and 44,
The light transmitted through the beam splitter 42 (about の of the returned light) is transmitted through the lens system 50 and the dichroic mirror 52, and the detection unit 5 of the FIA system as the detection optical system.
Go to 4. In the detection unit 54, the light from the dichroic mirror 52 is transmitted through a mirror 56 to an index plate 58.
An image of a wafer mark is formed thereon. This image and light from the index mark on the index plate 58 are passed through an imaging relay lens 60, a mirror 62, a relay lens 64, and a beam splitter 66, and each of the X-axis two-dimensional CCDs and the like is used. The images of the wafer mark and the index mark are formed on the imaging surfaces of the imaging device 68X and the two-dimensional imaging device 68Y for the Y axis.

In the focused state, the index plate 58 is arranged conjugate with the exposure surface of the wafer W with respect to the combined system of the objective lens 48 and the lens system 50. Further, the index plate 58 and the image sensors 68X and 68Y. The imaging surface of the relay lens system 60,
64 are conjugated to each other. The index plate 58 is formed by forming an index mark with a chrome layer or the like on a transparent plate, and a portion where an image of a wafer mark is formed remains a transparent portion. The index marks include an X-axis index mark serving as a position reference in a direction conjugate to the X direction on the wafer W and a Y-axis index mark serving as a position reference in a direction conjugate to the Y direction. The imaging devices 68X and 68Y capture images of at least one of the X-axis wafer mark and the Y-axis wafer mark and the image of the index mark, and process the imaging signals of the imaging devices 68X and 68Y, respectively. The X coordinate of the mark (relative position with respect to the index center) and the Y coordinate of the Y-axis wafer mark (relative position with respect to the index center) are obtained. Note that an illumination system for independently illuminating the index plate 58 is provided separately, but is omitted in FIG.

In this case, the objective lens 48 and the prism mirror 32 closer to the wafer W than the prism mirror 46 are called a common objective system. This common objective system is the FIA of the present embodiment.
It is commonly used for the alignment sensor and the focus sensor of the system. Common objective and FI after mirror 56
Each lens of the detection unit 54 of the A system is coaxially arranged along the optical axis AXa.

Next, the focus sensor constituting the sensor unit 16 will be described.

This focus sensor includes a light source 70 such as an LED (or laser diode) for the focus sensor,
Condensing lens 72, slit plate 74, dichroic mirror 38, lens system 40, beam splitter 42, prism mirrors 44 and 46, objective lens 48, lens system 5
0, dichroic mirror 52, light blocking plate 76 for pupil restriction
And a line sensor 78 composed of a one-dimensional CCD or the like.

Here, each component of the focus sensor will be described together with its operation.

The detection light M emitted from the light source 70 illuminates the slit plate 74 via the condenser lens 72. On the slit plate 74, the same number of focus detection patterns as the number of detection areas of the alignment sensor set within the observation field of view of the objective lens 48 are formed. The detection light M that has passed through the focus detection pattern of the slit plate 74 passes through the dichroic mirror 38, and then travels through the lens system 40 to the beam splitter 42. Here, as the detection light M, light in a wavelength band (for example, red light to near-infrared light) insensitive to the photoresist on the wafer W is used, and the wavelength selectivity of the dichroic mirror 38 is light. Fiber 30
The characteristic is set such that the light in the wavelength band used for position detection is reflected in the illumination light L3 from the camera, and the light in the wavelength band used for focus detection in the detection light M is transmitted. That is,
The light for position detection and the light for focus detection applied to the wafer W have different wavelength bands so that they do not adversely affect each other.

The light for focus detection reflected by the beam splitter 42 is applied to the prism mirrors 44 and 46 and the objective lens 4.
The light is irradiated onto the wafer W via the prism 8 and the prism mirror 32. The slit plate 74 includes the lens system 40 and the objective lens 4.
8, the image of the focus detection pattern in the slit plate 74 or an image obtained by defocusing the image is projected on the exposure surface of the wafer W. The light reflected on the exposure surface of the wafer W is reflected by the prism mirror 3
2. The beam returns to the beam splitter 42 via the objective lens 48, the prism mirror 46, and the prism mirror 44, and the light transmitted through the beam splitter 42 travels to the dichroic mirror 52 via the lens system 50.

The wavelength selectivity of the dichroic mirror 52 is different from that of the dichroic mirror 38 in that the optical fiber 3
0 through which the light for position detection emitted from the light source 70 is transmitted.
It has the characteristic of reflecting the light for focus detection emitted from the camera. Therefore, the light for focus detection reflected by the dichroic mirror 52 is projected onto the line sensor 78 and onto the wafer W via the outside of the pupil-limiting light-shielding plate 76 provided to break the telecentricity on the image side. An image (or a defocused image) of the focus detection pattern thus formed is re-imaged. That is, the exposure surface of the wafer W and the light receiving surface of the line sensor 78 are substantially conjugate with respect to the objective lens 48 and the lens system 50. As can be seen from this, in the present embodiment, strictly speaking, the condenser lens 72 and the slit plate 7
4. The light-shielding plate 76 and the line sensor 78 constitute a focus detection system.

FIG. 3 shows only the focus sensor in FIG. 2 for easy understanding. In FIG. 3, mirrors and the like for bending the optical path in FIG. 2 are omitted. As shown in FIG. 3, for example, three focus detection patterns 80A and 8 are provided in the slit plate 74.
0B and 80C (however, the pattern 80C is not shown in FIG. 3) are formed, and the images 81A, 81B and 81C of the focus detection patterns 80A, 80B and 80C by the lens system 40 and the objective lens 48 ( However, an image 81C is not shown in FIG. Then, the objective lens 48, the lens system 50, and the dichroic mirror 52 for the images 81A, 81B, and 81C.
And the images 82A and 82 re-imaged via the light shielding plate 76
B and 82C (the image 82C is not shown in FIG. 3) are projected onto the line sensor 78.

Since the images 81A, 81B and 81C of the focus detection pattern on the wafer W are arranged substantially on one straight line, the images 82A, 82B and 82C re-imaged on the line sensor 78. Are formed at different positions along a predetermined direction (this is referred to as a U direction). Here, the pupil-limiting light-shielding plate 76 shields a lower half area from the optical axis AXa in the U direction. In this case, the side from the objective lens 48 to the wafer W is telecentric, but the function of the light shielding plate 76 makes the side from the lens system 50 to the line sensor 78 non-telecentric. Therefore, when the wafer W is displaced parallel to the optical axis AXa (in the Z direction), the images 81A and 81 of the focus detection pattern on the wafer W
Although the positions of 1B and 81C do not change (the image is defocused), the images 82A and 82 on the line sensor 78 do not change.
The positions of B and 82C are shifted in the U direction. Utilizing this, the Z of the corresponding measurement point on the wafer W is determined from the amount of lateral displacement of the image on the line sensor 78 with respect to the reference position in the U direction.
The position in the direction (focus position) is detected.

Here, an example of the observation visual field on the wafer detected by the alignment sensor via the objective lens 48 will be described with reference to FIG. In FIG. 4, a circular area represents an observation field 84 of the objective lens 48 on the wafer W. Within the observation field of view 84, three detection areas 86A to 86C
Is set.

The detection areas 86A, 86B and 86C are the detection areas of the alignment sensor. A rectangular area 88 in the observation field of view 84 indicates an imaging range of the imaging elements 68X and 68Y. The detection area 86B indicates an area where an image is analyzed by the image sensor 68X, and the detection area 86C indicates an area where an image is analyzed by the image sensor 68Y. The detection area 86A indicates an area where both the image sensors 68X and 68Y perform image analysis in order to measure the registration (overlay) of the wafer W.

In FIG. 4, detection areas 86A to 86C virtually show wafer marks detected in the respective areas.

That is, in the detection area 86B, a line and space pattern (hereinafter referred to as "L / S pattern") 90X formed at a predetermined pitch in the X direction is detected, and in the detection area 86C, a predetermined pitch in the Y direction is detected. The L / S pattern 90Y formed by is detected. Further detection area 8
At 6A, a box-in-box type wafer mark 92 is detected. The wafer mark 92 includes an inner box 92b and an outer box 92a.

The L / S patterns 90X and 90Y and the marks 92 are respectively formed in the mark areas MAa and MAb between the plurality of shot areas SA on the wafer W, for example, as shown in FIG. Things.

In the measurement of the registration, the amount of positional deviation between the image of the outer box 92a and the image of the inner box 92b is obtained based on the image signals of the image sensors 68X and 68Y (see FIG. 2). It is performed by

FIG. 6 shows a detection area 86 A in the observation field of view 84.
9A to 9C show pattern images 81A to 81C for focus detection projected on each of FIGS. As shown in FIG. 6, the focus detection pattern images 81A to 81C projected on the detection areas 86A to 86C, respectively, are formed at a predetermined pitch in a direction crossing the X axis at 45 °. It is a light-dark pattern (multi-pattern).

In this case, since these image patterns have a shape that intersects with the wafer marks detected in the respective detection areas, the influence of the marks on the focus detection is reduced. The pattern for focus detection may be in the form of one slit pattern.

FIG. 7 shows focus detection pattern images 82A to 82A to be re-projected on the light receiving surface 78a of the line sensor 78 (see FIG. 3) corresponding to the focus detection pattern images of FIG.
82C is shown. In FIG. 7, light receiving elements are arranged on a light receiving surface 78a of the line sensor 78 along the direction of the arrow U, and pattern images 82A to 82C for focus detection are projected at different positions along the direction of the arrow U. Each of the pattern images 82A to 82C is a light / dark pattern formed at a predetermined pitch in the direction of arrow U,
The position of the pattern images 82A to 82C in the direction of the arrow U is obtained by processing the imaging signal read from the line 8 by the alignment control unit 18 (see FIG. 1).

The wafer mark 9 is obtained by determining the amount of displacement uA in the U direction from the reference position of the pattern image 82A.
The position shift amount in the Z direction of the area on the wafer W on which the wafer 2 (see FIG. 4) is formed is detected. By calculating the amount of displacement uB in the U direction from the reference position of the pattern image 82B, the amount of displacement in the Z direction of the region on the wafer W where the L / S pattern 90X (see FIG. 4) is formed is detected. The L / S pattern 90Y (FIG. 4) is obtained by calculating the amount of displacement uC of the pattern image 82C in the U direction with respect to the reference position.
The position shift amount in the Z direction of the area on the wafer W on which the reference (see) is formed is detected.

Here, an example of how to determine the reference position on the line sensor 78 will be briefly described.

In the case of an FIA type alignment sensor, for example, a reference mark plate FP (FIG.
Reference mark) is placed, and Z of the reference mark plate FP is placed.
While changing the position in the direction, the reference mark is
Take an image with X. Then, the imaging signal from the imaging element 68X is image-analyzed, and the reference mark plate FP is positioned in the Z direction so that the contrast of the image of the reference mark becomes highest. Therefore, the image 81B of the pattern for focus detection is projected onto the reference mark plate FP, and the pattern image 82B is projected onto the line sensor 78. The position of the pattern image 82B on the line sensor 78 at that time is represented by the pattern image 82B.
Of the reference position. The reference positions of the pattern images 82A and 82C may be determined in the same manner.

As the line sensor 78, a kind of two-dimensional image sensor in which a plurality of lines of pixels are arranged in parallel in the U direction is used. The position of the pattern image may be detected. Further, a cylindrical lens having a light condensing function in a direction perpendicular to the U direction in FIG. 7 may be arranged in front of the light receiving surface of the line sensor 78.

In FIG. 3, the line sensor 7 is moved from the wafer W of the combined system of the objective lens 48 and the lens system 50.
Assuming that the magnification (horizontal magnification) with respect to FIG.
1 / γ ^{2 of A to} uC is the amount of displacement in the Z direction from the focus position of the corresponding mark forming area on the wafer W. Therefore, when detecting the position of the wafer mark,
The position of the sample stage 14 in FIG. 1 is adjusted so that each of the positional deviation amounts becomes zero, and then the position is detected by the alignment sensor. This makes it possible to detect the position of the corresponding wafer mark with high accuracy in both the X direction and the Y direction of the FIA-based alignment sensor with accurate focusing. Moreover, the objective lens 48
A plurality of detection areas (for example, 86B and 86C) set in the field of view of each of the wafer marks (for example, 90
When X and 90Y) are located, the focus position can be simultaneously detected in the region on the wafer W where the respective marks are formed, so that there is an advantage that the position detection time of the wafer mark can be reduced.

In order to further improve the focusing accuracy, the auto-focusing may be performed by the servo system even when the position is measured by the alignment sensor.

Next, the operation of the exposure apparatus 10 according to the present embodiment configured as described above when performing exposure will be described. Here, for example,
Based on a measured value of the alignment mark position on the wafer W and a design value of the shot array as disclosed in Japanese Patent No. 44429 or the like, all shot array coordinates on the wafer W are calculated by a statistical operation using the least squares method. A case will be described in which the step-and-repeat exposure is performed by the EGA method in which each shot region is determined and the shot area is positioned at the exposure position based on the obtained values.

In this case, it is assumed that reticle alignment by a reticle microscope (not shown) has been completed.

First, the main controller 24 controls the drive system 2 so that the reference mark plate FP is located below the projection optical system PL.
The sample stage 14 is moved integrally with the XY stage 12 through the interface 2, and the output of the laser interferometer 20 at this time is stored in an internal memory (not shown). Next, in the main controller 24, the sample is integrated with the XY stage 12 via the drive system 22 so that the reference mark plate FP is located below the prism mirror 32 (the detection area of the alignment sensor in the sensor unit 16). The table 14 is moved, and the sensor unit 16 at this time is moved.
The output of the alignment sensor and the output of the laser interferometer 20 are stored in an internal memory. That is, the baseline measurement is performed in this manner. Note that the sequence of the baseline measurement is the same as that of the conventional exposure apparatus in this embodiment, and therefore, detailed description thereof is omitted.

Subsequently, in the main controller 24, the alignment marks (wafer marks) attached to a plurality of predetermined shot areas (sample shots) on the wafer W, for example, about 7 to 15, are displayed on the prism mirror. 32
While monitoring the output of the laser interferometer 20 so as to be positioned below the sensor interferometer 20 (detection area of the alignment sensor in the sensor unit 16), and sequentially positioning the sample stage 14 via the drive system 22, the output of the laser interferometer 20 The sequence proceeds to a sequence for sequentially detecting the position of each wafer mark based on the output of the alignment sensor.

At this time, before each measurement of the position of each wafer mark, the main controller 24 uses the focus sensor as described earlier to measure the position of each mark area in the Z direction before measuring the mark position. The shift amount (defocus amount) is detected, and it is determined whether the position shift amount exceeds a predetermined amount. Then, the mark position is not detected for the mark area where the Z direction displacement exceeds the predetermined amount, and the output of the alignment sensor and the laser interference are detected only for the mark area where the Z direction displacement is within the predetermined amount. The mark position is detected based on the output of the total 20. With this configuration, the measurement can be terminated when the position detection result of the alignment mark reaches a predetermined number (at least six or more predetermined numbers), so that the measurement can be completed as compared with the case where the measurement is performed on all sample shots. Measurement time can be shortened.

Next, the main controller 24 uses the measurement data of at least six alignment mark positions and the design data of the shot arrangement, which have been measured as described above, as described in JP-A-61-44429. All shot array coordinates on the wafer W are obtained by a statistical operation (so-called EGA operation) using the least square method disclosed in detail.

Thereafter, the main controller 44 determines each shot area (for example, 1) based on the above-described detection result of the alignment mark position and the above-described baseline measurement result.
(In the case of one chip shot, it corresponds to each semiconductor chip)
While controlling the position of the sample stage 14 so as to be sequentially positioned below the projection optical system PL, and performing an autofocus operation based on a focus position detection signal from a not-shown oblique incident light type focus sensor while illuminating. The opening and closing of a shutter (not shown) in the system IOP is controlled to repeat stepping and exposure of the sample table 18. In this manner, the overlap exposure is sequentially performed on each shot area on the wafer W by the step-and-repeat method.

As is clear from the above description, in the present embodiment, the above-described alignment sensor (30 to 6) is used.
6, 68X, 68Y) and the alignment control unit 18
Constitutes a mark position detecting means, and the above-mentioned focus sensors (32, 38 to 52, 70 to 78), the sample stage 14, the drive system 22, and the main controller 24 constitute a focusing means. 24 constitutes an arithmetic means.

As described above, according to the exposure apparatus 10 of the present embodiment, the sensor unit 16 in which the alignment sensor and the focus sensor are integrated is provided, and the detection area (area of the mark to be detected) of the alignment sensor is approximately equal to the sensor unit 16. The defocus amount of the minute area can be detected by the focus sensor, and the Z position of the sample stage 14 is adjusted based on the measurement result of the focus sensor, so that there is no defocus (focusing state). The position of the alignment mark (wafer mark) can be detected by the alignment sensor.

Prior to measuring the position of each wafer mark, the defocus amount of the mark area is measured, and the detection of the mark position is executed only in the mark area where the defocus amount is within a predetermined amount. When obtaining array data of all shot areas on a wafer, alignment marks formed in a region where the degree of unevenness is significantly larger than other portions, such as foreign matter such as dust adhered to the surface thereof The detection result of the position of the alignment mark formed in the area can be prevented from being included (excluded) in the basic data of the calculation of the shot array, and there is no adhesion of foreign matter or the like, and the defocus amount is within the predetermined range. Only the detection result of the position of the alignment mark is used as the basic data for calculating the shot array. Sequence of frequency is assumed accurate without errors due to foreign matter such as adhesion.

Then, based on the calculated array data of the shot areas, each shot area on the wafer W is sequentially positioned at a predetermined exposure position, and the image of the reticle pattern is sequentially exposed on a plurality of shot areas. Therefore,
Even when foreign matter or the like adheres to the wafer surface, it is possible to maintain the overlay accuracy with high accuracy while maintaining a higher throughput than in the case of the so-called die-by-die method.

Further, since the alignment sensor and the focus sensor which constitute the sensor unit 16 share the same objective lens 48, even if the optical characteristics of the objective lens 48 change, the optical characteristics of the objective lens 48 change. The influence of the focus detection due to the change is suppressed.

Further, when the reflected light from the wafer W is received in the detection system constituting the focus sensor, the light shielding plate 76 for restricting the pupil which breaks the telecentricity of the reflected light is provided. With this configuration, the Z-direction displacement (defocus amount) of the wafer mark area can be converted into a lateral shift amount of the re-imaged image and detected.

In the above-described embodiment, from the viewpoint of emphasizing throughput, a plurality of specific shots (sample shots) of wafer marks are determined in advance as detection targets, and the detection target marks are defocused by the focus sensor. Only when the amount is within a predetermined range, the position detection is sequentially performed in a focused state, and when the position detection result reaches a predetermined number, the array of shot areas is subjected to the least square method using the position detection result. The case where the calculation is performed by the used statistical processing has been described. However, the present invention is not limited to this. Position detection is performed sequentially in the in-focus state, and a mark in which the defocus amount (defocus amount) is within a predetermined range is detected. The sequence of the shot areas by using only the result may be calculated by statistical processing using the method of least squares. In this case, it is not necessary to determine whether the defocus amount is within a predetermined range at the time of position detection.
The control contents for detecting the mark position are relatively simple.

The imaging elements 68X and 68Y (see FIG. 2) constituting the FIA-based alignment sensor of the above-described embodiment image the same range in the field of view of the objective lens 48, and detect different areas in the entire imaging range. Although each image is analyzed, the imaging ranges of the imaging elements 68X and 68Y may be made different.

Further, for example, the detection area 86 shown in FIG.
If only the focus position in B needs to be measured,
Focus detection pattern 8 other than focus detection pattern 80B
Only a specific focus detection pattern can be obtained by shielding 0A and 80C with a shutter (not shown) or separately providing a light source for illumination for each of the focus detection patterns 80A, 80B and 80C and controlling light emission of these light sources. May be illuminated. Alternatively, a liquid crystal panel is provided in place of the slit plate 74, and only a desired position (a position corresponding to any of the focus detection patterns 80A, 80B, and 80C) of the liquid crystal panel is set in the light transmitting portion to thereby achieve a specific focus. Only the detection pattern may be generated.

Further, in the above-described embodiment, the case where the FIA type alignment sensor is used as the alignment sensor has been described. However, the present invention is not limited to this.
An IA (Laser Interferometric Alignment) type alignment sensor or an LSA (Laser Step Alignment) type alignment sensor may be used, or a plurality of alignment sensors may be used simultaneously by appropriately combining them. However, for example, the FIA system and L
When the IA system is used in combination with the IA system, a wavelength selection filter or the like for cutting laser light reflected from the wafer W when the LIA system is used is provided between the mirror 62 and the relay lens 64 in FIG. Devices such as arrangement are required.

[0082]

As described above, according to the first aspect of the present invention, it is possible to maintain a high throughput and to maintain a high overlay accuracy even when a foreign substance or the like adheres to the surface of the photosensitive substrate. There is an unprecedented excellent effect that it can be performed.

According to the second aspect of the present invention, in addition to the above-mentioned effects, even if the imaging characteristic of the objective optical system changes, the defocus of the alignment mark area is not affected by the change. There is also an effect that it is possible to accurately detect.

According to the third aspect of the present invention,
In addition to the effects of the first and second aspects of the present invention, the focus shift (displacement in the height direction) of the alignment mark area is converted into a lateral shift of the re-imaged image and detected with a simple configuration. There is also an effect that it becomes possible to do.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a schematic configuration of an exposure apparatus according to an embodiment.

FIG. 2 is a perspective view showing a configuration of a sensor unit in FIG.

FIG. 3 is a diagram schematically showing a configuration of a focus sensor in FIG. 2;

FIG. 4 is an enlarged plan view showing a detection area of an alignment sensor set in a field of view of an objective lens on a wafer.

FIG. 5 is an enlarged view showing an example of a region on a wafer where a positioning mark is formed.

6 is a diagram showing an image of a focus detection pattern projected on each of the detection areas of the alignment sensor shown in FIG. 4;

FIG. 7 is an enlarged view showing an image re-imaged on a line sensor.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 Exposure apparatus 14 Sample stand 16 Sensor unit (alignment sensor, focus sensor) 18 Alignment control unit 22 Drive system 24 Main controller 48 Objective lens (Objective optical system) 78 Shield plate (optical member) 86B, 86C Detection area 90X, 90Y L / S pattern for FIA system W Wafer (photosensitive substrate) R Reticle (mask) PL Projection optical system SA Shot area

Claims (3)

[Claims] An exposure apparatus for sequentially exposing an image of a pattern formed on a mask to a plurality of shot areas on the photosensitive substrate via a projection optical system while sequentially positioning the photosensitive substrate at a predetermined exposure position. Mark position detecting means for detecting a position of a positioning mark located in the detection region, the detection position having a detection area substantially equal to the positioning mark area provided in each shot area on the photosensitive substrate; Irradiating the detection area with a light beam and receiving the reflected light to detect a shift of the area with respect to the focal plane, and adjust the position of the photosensitive substrate in the optical axis direction based on the shift amount. Focusing means;
The shift amount detected by the focusing means is within a predetermined range;
Using only the detection result of the position of the alignment mark detected in the focused state by the mark detection unit, the plurality of shots on the photosensitive substrate are determined based on the detection result and the array data of the shot area in design. An exposure device comprising: an arithmetic unit configured to calculate an array of regions by statistical processing using a least squares method. 2. An exposure apparatus according to claim 1, wherein said focusing means irradiates said light beam and receives reflected light thereof through the same objective optical system as said mark position detecting means. apparatus. 3. The exposure apparatus according to claim 1, wherein the focusing unit includes an optical member that breaks telecentricity of the reflected light when receiving the reflected light.