JP2000035676A - Division sequential proximity exposure system - Google Patents

Abstract

(57) [Summary] [PROBLEMS] To provide an unprecedented highly accurate divided sequential proximity exposure apparatus. A pattern exposure unit irradiates a material to be exposed.
A work stage 2 on which a material to be exposed is mounted, a mask stage 1 on which a mask is mounted, a work stage feed mechanism 7 for positioning the work stage 1, and a gap adjusting means 6 between the mask and the material to be exposed. The proximity exposure apparatus includes mask position adjusting means 14 for adjusting the direction of the mask in the pattern plane with reference to the feed direction of the work stage feed mechanism 7.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a liquid crystal substrate, for example,
The present invention relates to a proximity exposure apparatus used for printing a mask pattern on a substrate such as a color filter for a liquid crystal display, and more particularly to a divisional sequential proximity exposure method suitable for proximity exposure of a mask pattern on a large substrate by a divisional sequential exposure method. Related to the device.

[0002]

2. Description of the Related Art In proximity exposure, a work is made of a light-transmitting substrate having a surface coated with a photosensitive agent, and the work is held on a work stage of a proximity exposure apparatus (hereinafter referred to as a proximity exposure apparatus). The gap (gap) between the mask and the mask attached to the mask stage on which the pattern is drawn is brought close to several tens μm to several hundreds of μm, and the mask is exposed on the work surface by irradiating parallel light from above the mask. In the exposure process of, for example, a color filter for a liquid crystal display, batch exposure has conventionally been common. However, recently, the screen size of the liquid crystal display has been increased, and it has become impossible to cope with the batch exposure, and it has become necessary to sequentially expose and connect a plurality of divided mask patterns. However, in the case of the divisional sequential exposure, there is a technical difficulty that the accuracy of alignment at the divisional boundary is strictly required.

[0003]

However, the conventional proximity exposure apparatus is basically constituted by a collective exposure method, and does not basically have a divided sequential exposure function for dividing a pattern and exposing and connecting the patterns. On the other hand, the conventional divisional sequential exposure apparatus does not have a highly accurate alignment function between the work stage and the mask, that is, an alignment function. There is an unsolved problem that the matching accuracy cannot be satisfied. FIG. 12 is a schematic diagram showing a deviation between the first shot exposure pattern and the second shot exposure pattern when, for example, the pattern formation of the first layer, for example, a black matrix, is performed by two-division sequential exposure without the alignment function. It is shown in a typical manner. Now, as shown in FIG. 3A, a mask M having a pattern P is arranged at a horizontal angle θ with respect to the running direction (feed direction) of the work stage.
Assume that the first shot exposure is performed in a state of being inclined only, and then the second shot exposure is performed after the workpiece W and the mask M are relatively moved and shifted as shown in FIG. As shown in (c), the divided pattern P by the first and second shots formed by exposure on the work W
Both ₁ and P ₂ are shifted by an angle θ with respect to the running direction (feed direction) of the work stage.

SUMMARY OF THE INVENTION The present invention has been made to solve such an unsolved problem of the prior art, and it is possible to realize a divisional and sequential proximity exposure with high precision which has not been achieved in the past, thereby increasing the size of a pattern and the accompanying mask. An object of the present invention is to provide a divided sequential proximity exposure apparatus that can cope with a tendency to increase costs.

[0005]

According to a first aspect of the present invention, there is provided an exposure means having an illumination optical system for pattern exposure for irradiating a parallel light to a material to be exposed. A work stage on which a material to be exposed is mounted, a mask stage on which a mask having a mask pattern is mounted, and a work stage and a mask stage which are relatively positioned so that the mask pattern faces a predetermined position on the material to be exposed. In a proximity exposure apparatus comprising: a stage feed mechanism for positioning the mask and a gap adjusting means for adjusting a gap between the mask and the surface to be exposed, the direction of the mask based on the feed direction of the stage feed mechanism. A mask position adjusting means for adjusting the position within the pattern plane is provided.

[0006]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is an exploded perspective view showing a stage configuration of a divided sequential proximity exposure apparatus according to the present invention, and FIG. 2 is an enlarged view of a mask stage portion. First, the structure will be described. In the figure, reference numeral 1 denotes a mask stage for holding a mask (M) having a mask pattern (P), and reference numeral 2 denotes a glass substrate (coated) on which a photosensitive agent is applied to print the mask pattern (P). A work stage 3 for holding the exposure material (workpiece) W is an illumination optical system as exposure means for pattern exposure. The illumination optical system 3 includes, for example, a high-pressure mercury lamp 31 which is a light source for irradiating ultraviolet rays, a concave mirror 32 for condensing the irradiation light, an integrator 33 disposed near the focal point of the concave mirror, and plane mirrors 35 and 36; An illumination optical system constituted by a spherical mirror 37 and an exposure control shutter 34 for shutting off irradiation light are provided. When the shutter 34 is opened at the time of exposure, light L emitted from the high-pressure mercury lamp 31 Through the optical path as described above, the surface of a mask (M) (not shown) held on the mask stage 1 is irradiated vertically.

The mask stage 1 has a mask stage base 11 and is fixedly supported on an apparatus base 4 via a mask stage support 12. The mask stage base 11 has an opening 111 at the center, and a mask holding frame 13 is slidably mounted in the X and Y directions surrounding the opening 111. FIG. 3 shows a sliding drive structure of the mask holding frame 13. As shown in FIG. 3A, which is a cross-sectional view taken along line III-III of FIG. 2, the mask holding frame 13 is inserted along the opening 111 of the mask stage base 11, and Rim 131
And a lower edge portion 1 is provided on the lower end face of the holding frame 13.
The upper and lower edges 132 and 132 are slidably engaged with each other so as to sandwich the mask stage base 11 therebetween.

[0008] Then, the mask holding frame 13 is moved on the X, Y plane on the upper surface of the mask stage base 11 to adjust the mask position with respect to the work surface. Direction drive 14x
And two Y-axis direction driving devices 14y.
The adjustment in the X-axis direction is performed by the former 14x, and the adjustment of the latter two 14
The y-axis direction and the θ-axis direction (swing around the Z-axis) are adjusted by y. Each mask holding frame driving device 14x, 14y
Are a drive actuator (electric actuator in this embodiment) 141 and a linear guide bearing (linear guide) 143 which is engaged via a pin support mechanism 142 at the tip of a rod 141r which expands and contracts in the axial direction. It is provided with. The slider 143 of the linear guide
s engages with the pin support mechanism 142, and the guide rail 143r is fixedly attached to the side surface of the mask holding frame 13, as shown in FIG. 3B which is a plan view.
A mask M on which a mask pattern is drawn is detachably held on the lower surface of the movable mask holding frame 13 via a vacuum suction device (not shown). In this holding state, the lower surface of the mask M is positioned below the lower surface of the mask stage base 11. Since the lower surface of the mask M protrudes downward as described above, when the work stage 2 is moved in the Y-axis direction, retraction in the Z-axis direction can be minimized.

A gap sensor 15 for measuring a gap between the mask and the work, and an alignment mark for determining the amount of displacement between the work W and the mask M are provided inside the two opposite sides of the mask holding frame 13. an alignment camera 16 as the detection means, all of which are movably disposed in the X direction via the moving mechanism 17. Specifically, the gap sensor / alignment camera moving mechanism 17 includes two holding frames 171 and 1 movably arranged in the X direction along the upper surfaces of the pair of left and right frames of the mask holding frame 13.
71, a linear guide 172 supporting one end thereof,
A driving actuator 173 for driving a slider (not shown) that moves on the guide rail 172r provided on the mask stage base 11, and
A holding frame 171 is fixed to the slider, and each gap sensor 15 and an alignment camera 16 are attached to the holding frame 171. By driving a driving actuator (in this case, a motor, a ball screw, and a linear guide) 173, the holding base 17 is driven.
The gap sensor 15 and the alignment camera 16 are moved in the X direction via the control unit 1.

The mask stage base 11 is provided with a masking aperture (shielding plate) 18 for shielding the end of the mask M as required, above the mask M in the opening of the mask holding frame 13. A masking aperture driving device 181 composed of a motor, a ball screw, and a linear guide is movable in the Y direction. The masking aperture (shielding plate) 18 has a large number of patterns arranged regularly as described in a method of transferring a large substrate, which was previously filed by the present applicant (Japanese Patent Application No. 10-42451). The rectangular exposure pattern is divided into an end pattern at both ends in the long side direction and a pattern on the inner side thereof, and the inner pattern is equally divided in the long side direction and divided and sequentially exposed. One divided pattern and the end patterns provided on both sides thereof are formed on one mask, and the end patterns on both sides of the mask are shielded when the inner pattern is divided and sequentially exposed using the mask. Used to do.

The work stage 2 is provided with a Z-axis feed base 6 serving as a gap adjusting means for adjusting a gap between the opposing surfaces of the mask M and the work W to a predetermined amount, and is disposed on the Z-axis feed base 6 so as to overlap therewith. And a work stage feed mechanism 7 for moving the workpiece in the Y-axis direction. The Z-axis feed base 6 includes a Z-axis coarse movement stage 62 supported by a vertical coarse movement device 61 erected on the apparatus base 4, and a Z-axis fine movement stage 64 supported thereon by a vertical fine movement device 63. It is composed of For example, a pneumatic cylinder is used for the vertical coarse moving device 61, and the Z-axis coarse moving stage 62 is moved up / down to a preset position without measuring a gap by performing a simple vertical operation.

On the other hand, the vertical fine-movement device 63 is a movable wedge mechanism that combines a motor, a ball screw and a wedge. That is, in the present embodiment, for example, the motor 631 is installed on the upper surface of the Z-axis coarse movement stage 62 to rotate the screw shaft 632 of the ball screw, and the ball screw nut 633 is formed in a wedge shape. The movable wedge mechanism 63 configured by engaging the slope of the nut 633 with the slope of a wedge 641 protruding from the lower surface of the Z-axis fine movement stage 64 constitutes a vertical fine movement device. When the screw shaft 632 is driven to rotate, the wedge-shaped nut 633 is guided by a guide (not shown) and moves in the Y-axis direction.
1 is converted into a high-precision vertical fine movement by the slope action. An up-down fine movement device comprising the movable wedge mechanism 63 is
Using two units, two units are arranged on one end side of the Z-axis fine movement stage 64 in the Y-axis direction, and one unit is arranged on the other end side, and each is independently driven and controlled. This allows
The vertical fine movement device 63 has a tilt function of finely adjusting the inclination with respect to the horizontal plane in addition to the function of finely adjusting the height of the Z-axis fine movement stage 64 to a target value while measuring the gap. I have.

In the case of performing the divided sequential exposure, the direction of the mask pattern of the mask M is deviated from the feed direction (Y-axis direction) of the work stage 2 even if there is no feed error of the work stage 2 at all. In addition, the seam between the patterns formed by the joint exposure is shifted and does not match. However, as described above, the mask M is sucked and held on the lower surface of the mask holding frame 13 via the vacuum suction device.
It is difficult to accurately match the direction of the mask pattern with the direction of movement (Y-axis direction) of the work stage 2 by the work stage feed mechanism 7 during the suction holding. Therefore, in the present embodiment, a table reference mark 75 having a cross shape, for example, is provided on at least two places on the upper surface of the work stage 2 (specifically, a work chuck installed on the stage).
Are formed as alignment marks, while a mask reference mark 76 (hereinafter, also referred to as a mask mark) corresponding to the table reference mark 75 is formed on the mask M as an alignment mark. The table mark 75 and the mask mark 76 are aligned by the mask position adjusting means 14 (i.e., the centers of the corresponding table mark 75 and mask mark 76 are substantially coincident in the XY plane). ), Mask holding frame 1
The position adjustment of 3 can be performed.

The principle by which the alignment of the first divisional patterns is made more accurate by the present invention is as follows. In the present invention, the Y-axis direction is determined as a direction parallel to the reflection surface of a mirror 741 for a laser interferometer 743 in the X direction described later. That is, the straightness standard in the Y-axis direction is determined by the mirror 741. When the table reference mark 75 and the mask mark 76 are aligned, the direction of the mask pattern of the mask M is not inclined with respect to the Y-axis direction, that is, the direction parallel to the reflection surface of the mirror 741. It is set to be. Therefore, the line connecting the centers of the two mask marks 76 in the mask pattern and the direction of the mask pattern, that is, the direction to be matched with the Y-axis direction (in the case of the rectangular pattern P in FIG. 11 and FIG. (The direction of the short side) is made equal to the angle between the line connecting the centers of the two table reference marks 75 and the reflection surface of the mirror 741.

In this embodiment, the reflecting surface of the mirror 741 for the laser interferometer 743 in the X direction is exactly defined by the line connecting the centers of the table marks 75 set at two places on the work stage 2. , And the relationship between the line connecting the centers of the mask marks 76 and 76 and the direction of the mask pattern was made the same. After the above, alignment with the mask M is performed with reference to the table mark 75, which is a work-side alignment mark fixed to the work stage 2, and exposure of a half pattern is performed first. By doing so, the shift between the first half pattern caused by the shift between the Y-axis direction and the direction of the mask pattern in the initial state and the second half exposure pattern formed after moving the work stage 2 can be prevented. Can be prevented.

Further, in the present embodiment, it is possible to correct the deviation between the first half pattern and the second half pattern caused by a feed error generated when the work stage 2 is moved by the work stage feed mechanism 7. This shift amount is sufficiently small compared to the shift amount between the Y-axis direction and the direction of the mask pattern 2 that can occur in the initial state described above. However, by correcting this shift amount, the first half can be more accurately adjusted. The other half can be matched.

The details of the work stage feed mechanism 7 and the correction of the feed error in the Y-axis direction will be described below.
The work stage feed mechanism 7 includes a Z-axis fine movement stage 6
4, two sets of linear guides 71 extending parallel to the Y-axis direction, a Y-axis feed base 72 attached to a slider (not shown) of the linear guides 71, and a Y-axis feed driving device 7.
The Y-axis feed base 72 is connected to a ball screw nut (not shown) screwed to a ball screw shaft 732 that is rotationally driven by a motor 731 of the feed driving device 73. The work stage 2 is mounted on the Y-axis feed base 72, and a mirror of a laser interferometer as feed error detecting means 74 for detecting a Y-axis feed error of the work stage 2 is provided. . That is, a laser mirror 741 extending in the Y-axis direction is provided along one side surface of the Y-axis feed base 72, and the other two mirrors 742 are provided on one end side of the Y-axis feed base 72 in the Y-axis direction. ing.

The feed error detecting means 74 includes a laser interferometer 743 for straightness detection, which is disposed opposite to the laser mirror 741 extending in the Y-axis direction and is supported by the apparatus base 4, and the two It is provided with two laser interferometers 744 for detecting the tilt and the distance in the Y-axis direction, which are respectively disposed opposite to the mirror 742 and supported by the device base 4. The detected feed error of the Y-axis feed base 72 and thus the work stage 2 is detected, and the result is sent to a correction control means (not shown).

Here, the feed error of the work stage 2 according to the present embodiment will be described. The feed error when the work stage 2 is fed by a predetermined distance in the Y-axis direction is an error of the feed position (distance). In addition, there may be straightness and inclination of the work stage surface. The straightness of the former is
The amount of deviation (in the XZ plane) at the feed end point between the Y axis passing through the feed start point and the actual feed travel direction straight line, and X
It comprises an axial component ΔX and a Z-axis component ΔZ. The latter inclination of the work stage surface is caused by yawing (rotation about the Z axis), pitching (rotation about the X axis), rolling (rotation about the Y axis), and the like when the work stage 2 moves. In yawing, the work stage is swung to the left or right while the work stage surface is horizontal, and is shifted by an angle θ with respect to the Y axis. In pitching, the work stage surface is inclined at a predetermined angle before and after in the traveling direction. Further, in rolling, the work stage surface is inclined left and right by a predetermined angle with respect to the horizontal. These errors occur due to, for example, the accuracy of a linear guide as a guide device.

Among these feed errors, the yaw and the straightness X-axis component ΔX can be known by the feed error detecting means 74. That is, yawing can be detected from the difference between the measured values of the two laser interferometers 744 and 744 in the Y-axis direction, and the straightness can be obtained by the result and the laser interferometer 743 in the X-direction. Also, the error in the Y-axis direction position is determined by the two laser interferometers 744 and 74 in the Y-axis direction.
4 can be obtained. Other, straightness Z
The axial component ΔZ, pitching and rolling can be known by three gap sensors 15.

In the divided sequential proximity exposure apparatus of this embodiment, the alignment between the alignment marks 75 and 76 is
The alignment camera 16 serving as the alignment mark detecting means described above can easily and accurately perform the operation. That is, as shown in FIG. 4, the basic configuration of the alignment mark detecting means is a mask mark 76 on the surface of the mask M held on the lower surface of the mask stage 1.
Is moved relative to the mask M by the focus adjustment mechanism 161 to adjust the focus. Specifically, the focus adjustment mechanism 161 is constituted by the linear guide 162, the ball screw 163, and the motor 164, and the guide rail 162r of the linear guide is mounted on the mask stage 1 (specifically, the holding base 171 of the moving mechanism 17). ), And vertically attached to the slider 162s.
The alignment camera 161 is fixed via 62t. The nut screwed to the screw shaft of the ball screw 163 is connected to the table 162t, and the screw shaft is driven to rotate by the motor 164.

In this embodiment, as shown in FIG. 5, an optical system 78 for projecting a table mark 75 from below the work chuck 8 provided on the work stage (2) includes a light source. 781 and condenser lens 78
2 and is arranged integrally with the Z-axis fine movement stage 64 in alignment with the optical axis of the alignment camera 16.
The optical system 7 is provided on the work stage 2 and the Y-axis feed base 72.
8 through holes corresponding to the eight optical paths.

Further, in the present embodiment, as shown in FIG. 6, the position of the surface of the mask M having the mask mark 76 (mask mark surface M _m ) is detected to prevent the alignment camera 16 from being out of focus. A best focus adjustment mechanism 9 for alignment images is provided. This best focus adjustment mechanism 9 uses the gap sensor 15 as a focus deviation detection means for an alignment camera 16 and a focus adjustment mechanism 161 which are the basic components of the alignment mark detection means shown in FIG. The measured value of the lower surface position of the mask measured by the above is compared with a focus position set in advance by a computer 92 to obtain a difference, and from the difference, a relative focus position change amount from the set focus position is calculated. The alignment camera 16 is moved by the focus adjustment mechanism 161 so as to adjust the focus.

By using the best focus adjustment mechanism 9, it is possible to adjust the focus of the alignment image with high accuracy irrespective of a change in the thickness of the mask M or a variation in the thickness. Note that the focus adjustment mechanism 161, the table mark projection optical system 78, and the focus adjustment mechanism 161 as shown in FIGS.
Various configurations such as the best focus adjustment mechanism 9 not only correspond to high precision of alignment of the first layer division pattern, but also contribute to high precision of alignment of the second and subsequent layers. In addition, the present invention is not necessarily limited to only the sequential exposure, and can be effectively applied to other exposures. If the thickness of the mask M is known, the best focus adjustment mechanism 9 in FIG. 6 may be omitted, and the focus adjustment mechanism may be moved according to the thickness.

Although not shown, the above-described divided sequential proximity exposure apparatus includes a bridge exposure position correction control means for performing alignment during bridge exposure to correct a positioning error of the mask pattern. The correction control unit corrects the plane deviation between the table reference mark 75 and the mask reference mark 76 detected by the alignment mark detection unit 77 by the mask position adjustment unit 14, and corrects the work W feeding direction (Y-axis direction) and the mask M This function is performed in a process after the first divided pattern is exposed and printed by matching the relative postures. That is, the feed error of the work stage 2 which occurs in the feed stage where the work is sent to the next area in order to connect and expose the second divided pattern is detected by the feed error detecting means 7.
4 and calculates a positioning correction amount for the splicing exposure based on the result. The calculation result is used as the mask position adjusting means 14 (and the vertical fine movement device 63 if necessary).
To correct the positional deviation.

Next, the operation of the present embodiment will be described. For example, RGB for color filter type large liquid crystal display
When a predetermined pattern is formed on a color filter, a black matrix pattern for partitioning between pixels is first formed on a glass substrate through steps of resist coating, oxygen blocking film coating, exposure, and development. On the substrate on which the black matrix pattern is formed, R (red), G
Individual patterns of the three primary colors (green) and B (blue) are formed for each color while repeating the same steps as those for forming the pattern of the black matrix. Here, the pattern to be formed first, that is, the pattern of the black matrix is the “first layer”, then any one of the three primary colors formed thereon is the “second layer”, and the other three primary colors are formed. Are referred to as “third layer”.

Now, a case where a glass substrate of a color filter for a large liquid crystal display is used as a work and a pattern of a first black matrix is formed thereon by a two-division sequential proximity exposure will be described. A mask M on which a first-layer pattern is drawn is previously mounted on the lower surface of the mask stage 1 by vacuum suction. Further, the work stage 2 is located near the forward limit in the Y direction and is lowered to the lowermost limit. (1) Gap Adjustment for Alignment First, the upper / lower coarse movement upper limit target position (for example, a mask) in which the work stage 2 is set in advance by driving the vertical coarse and coarse movement device 61 of the Z-axis feed base 6 which is a gap adjustment means. (Several mm from the surface of M). At the time of this coarse movement, the upper surface of the work chuck 8 by the gap sensor 15 (specifically, the glass test portion 8 fixed thereon)
The measurement of the gap (gap) between the (m upper surface) and the mask is not performed. Next, the vertical fine movement device including the three movable wedge mechanisms 63 of the Z-axis feed base 6 is driven. At the time of this fine movement, as shown in FIG.
The gap between the surface of the mask M having the workpiece 6 and the surface of the work chuck 8 having the work-side alignment mark 75 is measured by the gap sensor 15, and the measurement result is set in advance while being fed back to the movable wedge mechanism 63. The work stage 2 is raised accurately to the target gap amount at the time of alignment. As described above, according to the present embodiment, the vertical movement speed of the gap adjusting means is divided into two speeds, most of which is moved at a high speed by the vertical coarse and coarse movement device 61, and finally, the movable wedge mechanism 63 increases the speed. Since the movement is performed with precision, it is possible to simultaneously achieve the contradictory functions of speeding up the gap adjustment work and increasing the precision in the divisional sequential proximity exposure. It should be noted that the gap adjusting means of the present embodiment can be applied not only to sequential exposure but also to other types of exposure. (2) Alignment Adjustment In this way, when the work stage 2 and the mask M are brought close to each other, the running direction (Y-axis direction) of the work W and the relative position and orientation of the mask M are not correctly aligned. As shown in (a), when the mask M having the pattern P is inclined by the angle θ with respect to the work stage 2 (assuming that the direction of the work stage 2 and the direction in the Y-axis direction match). The table mark 75 provided on the work stage 2 side and the mask mark 76 on the mask M side are not aligned and are shifted. If the mask pattern is exposed and printed as it is, a deviation from the division pattern that is to be formed on the same work stage 2 thereafter occurs, and a black matrix pattern with high accuracy cannot be obtained after all. Therefore, an alignment operation (alignment) between the table mark 75 and the mask mark 76 is required. The alignment between the table mark 75 and the mask mark 76 is performed using the alignment camera 16 as the alignment mark detecting means, and at this time, both the marks 75 and 76 are focused.

In the first-layer division exposure according to the present embodiment, as described in (1), the table mark 75 is irradiated from below by the table mark projection optical system 78 as shown in FIG. Uses the gap sensor 15 used at the time of exposure. That is, the surface (the part to be inspected) 8 of the work chuck 8 having the table mark 75
By setting a gap on the surface of the mask M (mask mark surface M _m ) having the mask mark 76 and the mask mark 76, the alignment camera 16 serving as an alignment mark detecting means is provided on both the table mark 75 and the mask mark 76.
And the shift between the marks 75 and 76 is detected.

Normally, the thickness of the mask M is standardized, and is determined to be 5 mm, 8 mm, 10 mm, or the like.
The conventional alignment camera is not provided with the continuous focus adjustment mechanism 161 as in the present invention and has only one fixed focal point. There is a problem that the detection accuracy is reduced. However, in the case of the present invention, by setting in advance the focus positions of the alignment camera 16 corresponding to the various mask plate thicknesses as described above in the focus adjustment mechanism 161, the mask thickness due to the exchange of the mask M is set. and responsive to a significant change, it is possible to position the focus of the alignment camera 16 to the mask mark surface M _m is the best focus position.

In the present embodiment, the gap sensor 15 is used to detect the defocus, as described above. The actual lower limit position of the mask is measured by this, and the measured value is sent to the computer 92 and The relative focus position change amount from the focus position (target position) set on the lower surface of the mask (mask mark surface M _m ) is calculated, and the position of the alignment camera 16 is corrected by operating the focus adjustment mechanism 161 accordingly. By doing so, it is possible to adjust the focus shift due to, for example, variations in the mask thickness, and to guarantee the best focus for the mask mark 76.

When a deviation between the table mark 75 and the mask mark 76 is found by the alignment camera 16 whose focus has been properly adjusted as described above [FIG.
(A)], by instructing the mask position adjusting means 14 to drive the X-direction driving device 14x and the two Y-direction driving devices 14y, the posture of the mask holding frame 13 is corrected and both marks 75 and 76 are drawn. Correct alignment is performed as shown in FIG. Thus, the inclination θ between the mask M and the Y-axis direction (in the figure, the long-side direction and the Y-axis direction of the work W, the short-side direction of the mask M, and the short-side direction of the mask pattern P are respectively parallel. (Which indicates a certain case) is eliminated. (3) Work Insertion and First Exposure After the above alignment is completed, the work stage 2 is once lowered by the gap adjusting means 6 as needed. In this state, an interval between the mask stage 1 (for example, 60 m)
(approximately m), the work W is loaded by a work automatic supply device (not shown), and is held by the work chuck 8 on the work stage 2. After that, the gap adjusting means 6
Thereby, the clearance between the lower surface of the mask M and the upper surface of the work W is adjusted to a predetermined value required for exposure. The procedure is the same as the above (1) except that the gap sensor 15 measures the gap between the lower surface of the mask and the upper surface of the work.

When the work stage 2 is moved up and down by the gap adjusting means 6, the work stage 2 may slightly move even in the XY plane, though slightly. Therefore, for such a case, each of the laser interferometers 743, 744, and 74 after the completion of the alignment in (2).
4 is stored by the correction control means. If the position data after the gap adjustment is different from the stored data, the position of the mask M is corrected by the amount of change by the mask position adjustment means 14 so that there is no inclination between the direction of the mask M and the Y-axis direction. You can return to the state. Next, exposure is performed by operating the exposure control shutter of the exposure means 3, and the mask pattern P is printed on a predetermined position of the work W, thereby obtaining a _first divided pattern P1. (4) Stage Movement to Next Exposure Position Subsequently, the work drive ₂ of the work stage feed mechanism 7 is operated to move the work stage 2 in order to perform the joint exposure of the second divided pattern P2. As a result, the work W is moved by a predetermined distance in the direction of the arrow Y shown in FIG. At this time, in order to avoid interference with the work W, the work stage 2 may be lowered by a necessary amount in the Z-axis direction as needed. (5) when the gap adjustment and alignment adjustment above Y-direction feed, because feed error by factors mentioned above occurs, it albeit at when the next exposure second division pattern P ₂ slightly misaligned. For example,
Due to the yawing and the straightness error of the work stage 2 during feeding, the straightness Δx and the inclination angle θ ′ as shown in FIG.
Only from the normal position. Therefore, before performing the second exposure of the divided pattern P _2, detected by the error detection means 74 sends a feed error of the workpiece W, and sends the exposure position correction control means connecting the detection result. The linking exposure position correction control means calculates a positioning correction amount for the linking exposure based on the sent detection result, and based on the calculation result, the mask position adjusting means 14 (and pitching correction at the time of feeding, etc. In order to adjust the gap, the light is transmitted to the vertical fine movement device 63) to perform the alignment adjustment for correcting the positional deviation of the mask M. When the gap adjustment is performed, the alignment adjustment is performed based on yawing and straightness data in the subsequent state. (6) a second exposure then be performed stitching exposure, the second division pattern P ₂ is obtained that fixes position shift as shown in FIG. 7 (d).

According to the first embodiment of the present invention, it is possible to obtain the effect that it is possible to realize the divided successive proximity exposure of the first layer with higher accuracy than ever before. Further, the first layer exposure of the divided sequential exposure can be performed with the same accuracy as the second layer exposure using the global alignment of the conventional divided sequential exposure apparatus.

FIG. 8 is a diagram showing a second embodiment of the present invention. This is another embodiment of the optical system for projecting the table mark 75 from below the work chuck onto the mask alignment mark surface in a structure relating to the projection and imaging of the alignment mark in the divided sequential proximity exposure apparatus of the present invention. In addition, redundant description of the same components as those in the first embodiment is omitted. That is, the table mark projection optical system 78A of the present embodiment is
A light source 781, a condenser lens 782, and a table mark 75 are disposed below the table mark 75, and an image forming lens 783 is disposed above the table mark 75. The table mark projection optical system according to the first embodiment described above. 78 is different. The image forming lens 783 is arranged such that the table mark 75 is located at one focal point and the mask M is located at the other focal point.
The table mark 75 can be projected and imaged at the same level as the mask mark 76 by arranging such that the mask mark surface _Mm is positioned.

According to the second embodiment, the table reference mark 75 can be optically coincident with the mask reference mark 76 from the outside at the time of alignment so that the table reference mark 75 can be focused. And the mask reference mark 76 are not defocused from the detection optical system at all, and an ideal pattern can be detected. Thus, there is an effect that divisional sequential exposure of the first layer can be performed with higher accuracy. Note that the table mark projection optical system 78A can be applied to exposure other than the divided sequential exposure.

FIG. 9 is a diagram showing a third embodiment of the present invention. This is another embodiment of the best focus adjustment mechanism for the alignment image, particularly of the structure related to the projection imaging of the alignment mark in the divided sequential proximity exposure apparatus of the present invention, and is different from the above first embodiment. A duplicate description of the same components will be omitted. That is, the alignment image best focus adjustment mechanism 9A according to the present embodiment includes the alignment camera 16 and the focus adjustment mechanism 1
61 and a computer 92A.
In measuring the contrast of the image 76z of the pattern of the mask mark 76 on the mask mark surface M _m of the imaged mask M with the detected scan lines, differential waveform 7 of the image contrast
The state in which the wave heights D and D ′ of No. 7 are maximum is adjusted by the focus adjustment mechanism 1
At 61, the camera 16 is searched while moving. Thereby, there is an advantage that a camera position where the pattern image 76z is sharpest can be obtained, and alignment with even higher precision can be realized. Further, similarly to the second embodiment, there is no effect on the detection optical system with respect to an arbitrary mask thickness and a variation in the mask thickness, and an effect that high-precision detection can be performed for any mask thickness. Is obtained. This best focus adjustment mechanism 9A is not limited to the sequential exposure,
The present invention can be applied to other exposures.

In each of the above embodiments, the case where the first layer pattern is divided into two exposures has been described.
Of course, the present invention can be applied to a case where the operation is performed three or more times. In this case, at the time of the first exposure, a procedure of adjusting the gap between the mask M and the work W → correcting the inclination angle θ of the mask M with respect to the work W → exposure is performed. May be repeated as many times as necessary. Further, the division may be combined with the division in the X-axis direction instead of the division only in the Y-axis direction. In this case, an X-axis direction feed stage may be added.

In the above-described embodiment, only one mask M having the same pattern is used. However, the present invention is not limited to the case where the same pattern is repeated as described above. And Japanese Patent Application Hei 10-424
It is also possible to cope with a case where both end patterns as shown by 51 are provided.

When the pattern exposure of the first layer (that is, the black matrix) is completed as described above, a predetermined process such as development is performed, and a second layer, a third layer,... Can be Normally, color filters and the like are produced by line equipment, and the exposure apparatus uses a dedicated apparatus for the first layer, the second layer, and so on.

The exposure apparatus for the second and subsequent layers is, of course, provided with the apparatus of the present invention, that is, a mask position adjusting means using a table mark to detect and correct a feed error of the work stage to perform a continuous exposure. A divided sequential proximity exposure apparatus can be used. However, if the pattern of the mark used for the alignment of the second and subsequent layers is included in the exposure pattern of the first layer, the alignment based on the mark formed on the work is performed at the time of the exposure of the second and subsequent layers. In this case, the exposure apparatus for the second and subsequent layers may be a conventional exposure apparatus having a function of performing alignment between the alignment mark on the work and the alignment mark on the mask.

However, in such a conventional exposure apparatus, as described above, a mechanism for high-precision matching of the joint between the first divisional patterns is not considered. By producing with a line facility using the present invention as the first layer exposure apparatus, highly accurate joint alignment can be realized. In other words, by using the mask position adjusting device using the table mark as the first-layer exposure device, the work stage feed direction and the mask pattern direction can be aligned, and joints can be aligned. Becomes Then, at that time, by detecting a feed error of the works date and correcting the error, it is possible to achieve a more accurate joint alignment.

Hereinafter, as a fourth embodiment, an alignment adjustment in the case where the second-layer exposure is performed by the divided sequential proximity exposure apparatus of the present invention to which a function dedicated to the second-layer exposure is added will be described. When performing joint exposure of the second and subsequent layers using a conventional exposure apparatus, when the work table is moved in the Y direction toward the exposure position of the second shot after the work alignment of the exposure of the first shot is completed. In
The second shot is aligned after relying on the work table feeding accuracy or sending the work W as much as possible with the work table W being mounted on the work table in accordance with the work table feed error (inclination in the Y-axis traveling direction). The method is taken.

However, in such a method, when the work alignment mark of the second shot is detected, the work W when the work W is loaded onto the work table and mounted thereon is detected.
A table feed amount y (size of the inclination angle theta _w) of inclination accuracy
Depending on the size of the workpiece, the work alignment mark at the position of the second shot may not be in the field of view of the alignment camera, and it takes time to capture the work alignment mark in the field of view, and it takes a long time for the work alignment. And in the worst case, the work alignment of the second shot becomes impossible.

The present embodiment can solve such a problem in the exposure of the second and subsequent layers, and will be specifically described with reference to FIGS. 10 (a) to 10 (d). As the apparatus, the above-described divided sequential proximity exposure apparatus of the present invention is applied. The description up to the second shot has been described, but the same applies to subsequent shots.

First step [FIG. 10 (a)]: This is a stage of preparation for the second-layer exposure. The mask M for the second layer exposure is mounted on the mask stage 1. The mask marks 76, 76 of the mounted second-layer exposure mask M are aligned with the centers of the visual fields A, A of the respective alignment cameras 16. Then, before sending the work stage 2 to the position of the second shot, the actual feed direction of the work stage 2 (Y ′ direction)
Of the field of view A of the alignment camera 16 with respect to
Axis line connecting the centers of A) X _A orthogonality deviation angle θ _m
Obtained in advance (hereinafter referred to as mask misalignment angle theta _m).

The mask shift angle θ _m is a shift angle between the Y axis of the mask M and its actual relative feed direction Y ′.
The tilt angle θ between the running direction of the Y axis and the table (alignment) mark 75 during the first exposure (see FIG. 7)
This is the same as replacing the table mark 75 with the camera view A in that case. Later Specific Determination examples of mask misalignment angle theta _m.

Second step [FIG. 10 (b)]: The work W after the first-layer exposure is put on the work stage 2 and mounted. The work W is usually mounted in a state where it is relatively inclined in the clockwise (plus) or counterclockwise (minus) direction with respect to the actual feed direction Y ′ of the work stage 2. FIG. 10B shows a state immediately after the work W is mounted. The inclination angle θ _W of the work W with respect to the feed direction Y ′ is defined by the axis XW connecting the centers of the cross-shaped work alignment marks 77 (hereinafter referred to as the work alignment axis) _XW and the actual feed direction Y ′ of the work stage 2. angle between the axis X _Y orthogonal (hereinafter referred to as workpiece inclination angle theta _W) is. In this case, the work alignment mark 77
And there is still displaced from the mask mark 76, its displacement angle (mask tilt angle) alpha corresponds to the angle between the central axis X _A of the workpiece alignment axis X _W and the camera field of view A. That is, the work inclination angle θ _W in this case is θ _W = θ
_m + (minus α), and the deviation angle θ _{m of the} mask M
The angle becomes smaller in the counterclockwise direction by the shift angle α between the mask M and the work W.

Third step [FIG. 10C]: This is the first shot of the second-layer exposure. Alignment of the mounted work W and the mask M is performed to eliminate the displacement between the two. That is, the mask M is moved with respect to the work W to eliminate the displacement. Specifically, the mask M is rotated counterclockwise around the Z axis by the operation of the mask position adjusting means 14 to align the square mask mark 76 with the cross work alignment mark 77. The turning angle of the mask M, that is, the mask inclination angle α at that time is stored in the storage device of the exposure position correction control means. FIG. 10C shows a state at the time of completion of the alignment between the work alignment mark 77 and the mask mark 76 of the first shot in the second-layer exposure. After that, the first shot is exposed.

Fourth step (FIG. 10D): After the completion of the exposure of the first shot, the work stage feed mechanism 7 sends the work stage 2 to the exposure position of the second shot. In the feed operation, the feed amount (distance) y is measured by the laser interferometer 744 that detects the feed in the Y-axis direction. Next, the second shot alignment is started by rotating the mask M counterclockwise by the work tilt angle θ _W , but before that, the position of the mask M is corrected by the mask position adjusting means 14 of the mask stage 1. I do. Specifically, the feed amount (distance) y of the work stage 2 and the first
Using the mask shift angle θ _m obtained in the step and the mask tilt angle α obtained in the third step, the correction amount ΔX of the mask stage 1 in the X-axis direction is calculated by the following equation.

ΔX = y · sin (θ _m + α) The mask stage 1 is moved in the X direction by the distance of ΔX to correct the position. By this mask position correction, the cross work alignment mark 77 and the square mask mark 76 for the second shot always enter the field of view of the alignment camera 16. Therefore,
In the detection of the work alignment mark after the second shot, the magnitude of the inclination angle θ _W when the work is mounted,
No alignment error occurs regardless of the direction feed amount. Moreover, the takt time of the device is greatly reduced.

Further, theoretically, in the work alignment for the second and subsequent shots, the mask stage is moved to the alignment completed position when it enters the field of view by accurately performing the above calculation and its correction. Alignment of shots is no longer necessary (that is, the alignment is automatically performed by operating the mask position adjusting means 14 in accordance with a command from the correction control means, and two or even n shots can be exposed with one alignment). Thus, global alignment exposure that does not require work alignment for the second and subsequent shots is possible,
The tact time of the apparatus can be further reduced and the accuracy can be improved.

In the fourth embodiment, in detecting the work alignment marks 77 for the second and subsequent shots in the second-layer exposure, the actual feed direction Y of the work stage 2 determined in advance during the first-layer exposure is determined. ”And the position of the work is recognized from the result of the work alignment of the first shot at the time of the second layer exposure, and the position of the mask stage is corrected. Has been described in such a way that is always captured in the field of view of the alignment camera, but such a control method is not necessarily applied only after the second-layer exposure. By previously knowing the actual feed direction Y 'of the work stage 2 in a test manner, the present invention can be applied to the case of the inclination angle θ of the table alignment mark for the first-layer exposure and the inclination θ of the running direction of the work stage. .

Here, the mask shift angle θ_mHow to find
An example will be described. FIG. 11A shows the second-layer exposure.
This is the state of the first shot. That is, FIG.
As in (a), the mask alignment mark of the mask M
(□) 77_A, 77_BEach alignment card.
Align with the center of the visual field CA, CB (marked by ○) of the camera. this
At right angles to the actual running direction Y 'of the work stage
Axis X _YCenter line of sight X of camera_ADeviation angle θ_m
Is measured in advance by the following method.

The work stage is sent and the mask M is
Move relatively to the position of the shot. Since the actual feed direction is the Y ′ direction, the work alignment marks 77 _AI and 77 _{BI at} the position I of the first shot are Y ′
It moves to the position II along the direction and becomes the work alignment marks _77AII and _77BII (FIG. 11B).

The measurement standard is a line L connecting 77 _BI- 77 _BII.
s, and the workpiece W is set such that the reference line Ls is in the X direction.
Make a horizontal turn. Then, a straight line 77 with respect to the reference line Ls
_BI -77 _AI and 77 _BII -77 _AII tilt angles θ _m1 , θ _m2
[FIG. 10 (c)]. θ _m = (θ _m1 + θ _m2 ) / 2 = (tan ⁻¹ g ₁ / G ₁ + tan ⁻¹ g ₂ / G ₂ ) / 2.

Incidentally, in order to improve the accuracy, the processing may be performed up to n shots and the average may be calculated. θ _m = (θ _m1 + θ _m2 +... + θ _mn ) / n In the above embodiments, the mask stage 1 is fixed to the apparatus base 4 by the mask stage support 12 and only the work stage 2 has a gap. Although the structure for raising and lowering by the adjusting means 6 has been described, the present invention is not limited to this. For example, the mask stage support base 12 may be configured by a cylinder and the mask stage 1 may be raised and lowered. In this case, the Z-axis coarse movement stage 62 having the vertical coarse movement device can be omitted.

Further, in each of the above embodiments, the work stage is configured to be movable in the Y-axis direction. Alternatively, the mask stage and the exposure unit may be configured to be movable in the Y-axis direction.

[0058]

As described above, according to the present invention,
There is an effect that high-precision and high-speed divisional sequential proximity exposure that has not been achieved in the past can be realized.

[Brief description of the drawings]

FIG. 1 is an exploded perspective view showing a stage configuration of a divided sequential proximity exposure apparatus according to a first embodiment of the present invention.

FIG. 2 is an enlarged view of the mask stage portion.

3A is a cross-sectional view taken along the line III-III of FIG. 2, and FIG.

FIG. 4 is a side view of the basic structure of the alignment camera and its focus adjustment mechanism.

FIG. 5 is a configuration diagram of a table mark irradiation optical system according to the first embodiment.

FIG. 6 is a configuration diagram illustrating a focus adjustment mechanism for an alignment image according to the first embodiment.

FIG. 7 is an explanatory diagram of a procedure for adjusting a deviation between a table mark and a mask mark in the first embodiment.

FIG. 8 is a schematic diagram of a table mark irradiation optical system of a divided sequential proximity exposure apparatus according to a second embodiment of the present invention.

FIG. 9 is a schematic diagram of a focus adjustment mechanism of a divided sequential proximity exposure apparatus according to a third embodiment of the present invention.

10A and 10B are explanatory diagrams of a procedure for adjusting a shift between a work alignment mark and a mask mark after a second shot in a second-layer exposure according to a fourth embodiment of the present invention, wherein FIG. Schematic diagram showing alignment after loading,
(B) is a schematic diagram showing the state immediately after the workload, and (c) is the first view.
FIG. 9D is a schematic diagram showing a state of completion of the alignment of the shot, and FIG. 9D is a schematic diagram showing a state after moving to the alignment of the second shot.

[11] a view for explaining how to determine the mask misalignment angle theta _m applicable to the present invention, (a) is a plan view showing a first shot of the state of the second layer exposed, the (b) the mask M Second
A plan view showing a state of being relatively moved to the position of the shot,
(C) is a plan view showing a state at the time of measuring the mask shift angles θ _m1 and θ _m2 .

FIG. 12 is a schematic diagram showing a shift between an exposure pattern of a first shot and an exposure pattern of a second shot when pattern formation of a first layer is performed by sequential exposure in two without an alignment function; (A) is the first shot exposure in a state where the mask M is inclined by the horizontal angle θ with respect to the workpiece W, and (b) is the second shot exposure performed after the workpiece and the mask are relatively moved next. (C) shows the displacement error of the divided patterns P ₁ and P ₂ due to the first and second shots formed on the work W by exposure.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Mask stage 2 Work stage 3 Exposure means 6 Gap adjustment means 7 Work stage feed mechanism 14 Mask position adjustment means 16 Alignment mark detection means 74 Feed error detection means 75 Table reference mark 76 Mask reference mark

Claims (1)

[Claims] 1. An exposure means having an illumination optical system for pattern exposure for irradiating a parallel light to a material to be exposed, a work stage for mounting the material to be exposed, and a mask for mounting a mask having a mask pattern A stage, a stage feed mechanism that relatively positions the work stage and the mask stage so that the mask pattern faces a predetermined position on the exposure target material, and a gap between opposing surfaces of the mask and the exposure target material. A proximity exposure apparatus comprising: a gap adjusting means for adjusting; and a mask position adjusting means for adjusting a direction of a mask in a pattern plane with reference to a feeding direction by the stage feeding mechanism. apparatus.