JPH09258234A - Sticking device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To build a light source for distinctly photographing alignment marks with a television camera into a floating mechanism without occurrence of a hindrance in a floating function at the time of aligning upper and lower members. SOLUTION: A lower sucking means 20 consists of a vacuum sucking table 22 and a spherical part 23 of a projecting spherical shape disposed successively at this vacuum sucking table 22. This spherical part 23 fits onto a spherical seat 24, thereby, the floating mechanism is constituted. A light transmission member 33 which transmits the illumination light from a light source lamp 32 is disposed in order to illuminate the sections of alignment marks M1 , M2 from below at the vacuum sucking table 22 so as to face the flank of the vacuum sucking table 22. The vacuum sucking table 22 is provided with a light guide 34. The incident end 34a of the light guide 34 faces the exit end 33b of the light transmission member 33. The exit end 34b faces the lower position of the alignment mark M1 of a TFT substrate 1 on the front surface of the vacuum attraction table 22.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention positions two upper and lower members relative to each other and makes one of the upper and lower members strictly follow the other so that a predetermined gap is formed therebetween. The present invention relates to a bonding device.

[0002]

2. Description of the Related Art For example, a color TFT (Thin Film Transistor) is used as a liquid crystal panel in a liquid crystal display.
There is a liquid crystal panel. This type of liquid crystal panel is a TFT
A CF (Color Filter) substrate is assembled on the substrate so as to face each other with a liquid crystal interposed therebetween. For this, first T
An ultraviolet curable sealing material is supplied to a predetermined portion of the surface of the FT substrate (or CF substrate), the other substrates are superposed on each other, and the sealing material is compressed to form a predetermined gap between the two substrates. Stack them as you have them,
Then, the sealing material interposed between the TFT substrate and the CF substrate is irradiated with ultraviolet rays to cure the sealing material so as to fix the two substrates in a bonded state. Liquid crystal is injected into the inside and the inlet is closed.

In the liquid crystal panel manufactured as described above, in the process of bonding the TFT substrate and the CF substrate together, these two substrates are first sucked and held by the vacuum suction means, and the two substrates are held between them. Bonding is performed by applying a predetermined pressing force. In the step of bonding the substrates, at least one of the substrates is in the XY direction (vertical and horizontal directions) and the θ direction (rotational direction) in the horizontal plane in order to prevent displacement between the substrates.
The XYθ stage capable of position correction is used to detect the positional deviation of the both substrates in the XYθ directions by using the image recognition means, and the two substrates are accurately aligned to each other by applying a pressing force. It has been conventionally practiced to bond the.

In order to align the two substrates, two or more alignment marks are provided at predetermined positions on each substrate, and the alignment marks of the two substrates are aligned to align them. be able to. That is, these alignment marks are photographed by a television camera, the image signal from the television camera is taken into the image recognition means, the positional deviation of both substrates in the XYθ directions is detected by calculation, and XYθ is calculated based on the servo signals from the positions. By adjusting the position of the stage,
Both substrates can be aligned. Therefore, in order to be able to capture the alignment marks of both substrates with a TV camera, at least both substrates should be made of transparent members, and one vacuum suction means facing the TV cameras should also be made of transparent members. There must be.

Here, even if both substrates and one of the vacuum suction means are formed of transparent members, both alignment marks are made clear to the extent necessary or preferable for image processing by the television camera. You cannot always shoot. In particular, since one alignment mark is photographed through at least the other substrate and the vacuum suction means, the alignment mark may not be clearly photographed due to insufficient light amount. Therefore, it is necessary to provide a light source below this one alignment mark, irradiate illumination light from this light source in the direction of the television camera, and perform photography by the television camera under this illumination.

[0006]

By the way, when the two substrates are bonded together, the gap between them must be controlled extremely strictly, and the gap size is made uniform so that the gap size does not vary partially. May have to be. Thus, in order to perform fine gap management, it is necessary to accurately copy one substrate to the other substrate. Generally, various floating mechanisms are used as the copying mechanism. That is, when one side is held by the floating mechanism and brought into contact with the other member, the floating member can be made to follow the other member. However, when the light source is arranged to face the TV camera when the alignment mark is photographed, either the TV camera or the light source must be provided in the floating mechanism. Then,
There is a problem that it is not possible to perform accurate floating, and it becomes impossible to strictly adjust the gap due to uniform pressing force.

The present invention has been made in view of the above points, and it is an object of the present invention to provide an illumination light for clearly photographing an alignment mark with a television camera without disturbing a floating function. It is to be able to incorporate the irradiating part of the device into the floating mechanism.

[0008]

In order to achieve the above-mentioned object, the present invention comprises a lower member mounted on a vacuum suction portion of a floating member, and an upper member arranged to face the lower member. The lower member, the upper member, and the vacuum suction means are formed of a transparent or translucent member in order to perform alignment based on the alignment marks provided on each, and a television for image recognition is provided above the vacuum suction means. A camera is installed, and the vacuum suction part is equipped with a light guide made of an optical fiber whose emission end face of illumination light faces the lower part of the position where the alignment mark of its lower member is located, and whose incident end face faces the side surface. The feature is that the light source is arranged so as to face the incident end face in a non-contact manner.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION Both upper and lower members are sucked by vacuum suction means so that the upper member and the lower member face each other. In this state, using a TV camera for image recognition,
The alignment mark formed on both members is photographed to detect the positional deviation. Then, the positions of at least one of the upper member and the lower member are adjusted in the X, Y, and θ directions to align the two members.

The upper member is fixedly held by vacuum suction means. On the other hand, the lower member is in a floating state.
For this purpose, the vacuum suction portion is connected to the floating member. As the floating member, for example, a spherical portion can be connected to the vacuum suction portion, and the spherical portion can be fitted to a spherical seat to form a vacuum member. The lower member is vacuum-sucked to the suction unit. Then, in order to float the lower member, an air supply / discharge port is opened in the spherical seat, and pressurized air is supplied from this air supply / discharge port, so that the vacuum suction device in which the spherical portion and the lower member are installed. The part floats.

When aligning the upper and lower members,
Pressurized air is supplied from the air supply / discharge section to float the spherical section and the vacuum suction section connected to the spherical section. In this state, the upper and lower members are brought into contact with each other by lowering the upper member or raising the lower member. Since the lower member is in a floating state, the lower member surely follows the upper member, and the parallelism of both members is secured. Then, by making the air supply / discharge portion have a negative pressure, the spherical portion is adsorbed and fixed to the spherical seat, and a predetermined pressing force is applied to either the upper or lower member. As a result, both members are pressed while maintaining parallelism. Therefore,
By adjusting this pressing force, it is possible to perform the bonding in a state in which the gap between both members is accurately managed.

In order to detect the positional deviation between the upper member and the lower member, a television camera is used to photograph the alignment mark. Here, the upper member and the lower member are transparent or semi-transparent members, and the vacuum suction means for vacuum-sucking the upper member is a transparent or semi-transparent member so that the alignment mark can be photographed. To form. However, even if the alignment mark of the lower member is transparent, it is photographed by the television camera through the upper member and the vacuum suction means, so a clear image cannot be obtained unless illumination is performed from below. The vacuum suction part to which the lower side member is attached is connected to the spherical part that constitutes the floating member.When the spherical part is levitated, not only contact the spherical seat but also any fixed side member. Instead, it must be kept in a completely floating position.

In consideration of this point, the illumination mechanism from below is composed of a light guide made of an optical fiber and a light source. The light guide is integrated into the vacuum suction part, and the emission end of the illumination light is on the surface where the lower member is installed.
The alignment mark is arranged at the position. Further, the incident end of this light guide is arranged on the side surface of the vacuum suction section. Then, a light source is arranged opposite to the incident end of this light guide. Thereby, the light from the light source is transmitted through the light guide, and the illumination light is emitted toward the television camera from below the alignment mark.

With this structure, the spherical portion does not need to be connected to or abut on the fixed side member, that is, the floating portion does not have any trouble, and the necessary portion can be provided. Can be illuminated. As a result, both alignment marks can be clearly photographed by the television camera, and therefore image recognition can be accurately and reliably performed.

A light source lamp is required at least as a light source arranged opposite to the light guide provided in the vacuum suction portion, but a light collecting member may be arranged in order to collect light more efficiently. The light source lamp and the light condensing member may be directly opposed to the light guide, but in order to accurately oppose the incident end and facilitate position adjustment,
It is preferable that the light from the light source lamp is once passed through a light guide member made of a flexible optical fiber, and the light guide member is made to face the optical fiber on the vacuum suction side.

[0016]

Embodiments of the present invention will be described below with reference to the drawings. First, FIGS. 1 and 2 show the structure of a liquid crystal panel. In the figure, 1 is a TFT substrate, 2 is a CF substrate, and these TFT substrate 1 and CF substrate 2 are both transparent substrates. The substrates 1 and 2 are attached to each other via an ultraviolet-curable sealing material 3. Here, a spacer 4 is mixed in the sealing material 3, and the TFT substrate 1 and C
The spacer 4 is separated from the F substrate 2 by the thickness of the spacer 4, and the liquid crystal is injected into the gap G formed thereby.

When the TFT substrate 1 and the CF substrate 2 are attached to each other, it is necessary to align their positions in the XYθ directions. For this reason, the TFT substrate 1 and the CF substrate 2 are provided with alignment marks M ₁ and M ₂ at a plurality of locations, for example, at two locations. By aligning these alignment marks M ₁ and M ₂ , The position adjustment in the X, Y, and θ directions is performed. Also, T
The FT substrate 1 and the CF substrate 2 are bonded so as to have a gap G. Liquid crystal is injected into this gap G, but in order to make the thickness of this liquid crystal uniform, the gap G
It is necessary to control extremely accurately so that the dimension is uniform, for example, several μm or less, preferably 1 μm or less. In order to secure this gap G, a spacer 4 is interposed in the seal material 3. Then, in order to strictly control the gap G between the two substrates 1 and 2, a uniform pressure is applied from above and below by a pressing mechanism.

FIG. 3 shows the structure of a bonding device for the TFT substrate 1 and the CF substrate 2. A vacuum suction mechanism is used to bond the TFT substrate 1 and the CF substrate 2 together. Reference numeral 10 is an upper suction means, and 20 is a lower suction means.

The upper suction means 10 is movable in this embodiment, and has a suction plate 12 having a large number of suction holes 11, and a drive shaft 13 is connected to the suction plate 12. The drive shaft 13 is connected to at least a robot (not shown) capable of moving up and down. The upper suction means 10 may be the fixed side, the lower suction means 20 may be the movable side, or both may be movable. The drive shaft 13 is a hollow shaft, inside which a suction passage 13a communicating with a suction source is formed, and a lower end portion thereof is a holder portion 13b having an enlarged diameter. A transparent member 14 is fitted in the holder portion 13b of the drive shaft 13, and a negative pressure chamber 15 communicating with the suction hole 11 is formed between the transparent member 14 and the suction plate 12, and the suction passage is formed. 13 a is open to this negative pressure chamber 15.

On the other hand, the lower suction means 20 has a vacuum suction portion 22 having a large number of suction holes 21, and this vacuum suction portion 2
A spherical portion 23 having a convex spherical shape is connected to 2. The spherical portion 23 is arranged so as to fit on a spherical seat 24 having a concave spherical surface shape with the same radius of curvature as the spherical portion. A negative pressure chamber 25 is formed between the spherical portion 23 and the vacuum suction portion 22, and the suction hole 21 communicates with the negative pressure chamber 25. A suction nozzle 26 is opened in the negative pressure chamber 25 in order to make the inside of the negative pressure chamber 25 negative. The suction nozzle 26 includes a spherical seat 24 and a spherical portion 2 from the bottom.
3 is extended and the tip part is opened to the negative pressure chamber 25. For this reason, the spherical portion 23 and the spherical seat 24 are provided with through holes 23a and 24a at their central portions. The diameter of these through holes 23a and 24a is slightly larger than the outer diameter of the suction nozzle 26, so that the suction nozzle 26 is inserted into the through holes 23a and 24a in a substantially loosely fitted state.

The spherical seat 24 has one or a plurality of air supply / exhaust holes 2
7, an air passage (not shown) is connected to the air supply / exhaust hole 27, and the air passage is selectively connected to a supply source of compressed air or a negative pressure source. It Further, the air supply / exhaust hole 27 is opened in the circumferential groove 28 on the surface of the spherical seat 24, and when pressurized air is supplied to the air supply / exhaust hole 27, the spherical portion 23 floats above the spherical seat 24. As a result, the vacuum suction portion 22 connected to the spherical portion 23 is provided.
Becomes a floating state. On the other hand, when the air supply / exhaust hole 27 is connected to the negative pressure source, the spherical portion 23 is fixed to the spherical seat 24 by suction.

By the way, when the TFT substrate 1 and the CF substrate 2 are bonded together, it is necessary to detect the positional shift between the substrates 1 and 2 and perform the positioning. For this purpose, the relative displacement between the substrates 1 and 2 is detected by using the image recognition means. Therefore, a number corresponding to the number of each alignment mark (two locations in this embodiment) at the upper position of the drive shaft 13,
That is, two TV cameras 30 are arranged. In order to secure the field of view of the TV cameras 30, the suction plate 12 is formed of a transparent member such as quartz glass, and the TFT substrate in the holder portion 13b of the drive shaft 13 is provided. 1 and C
A through hole 13c is formed at a position above the alignment marks M ₁ and M ₂ of the F substrate 2, and the holder portion 13b is formed.
A transparent member 14 is provided inside. Therefore, the television camera 30 photographs the alignment marks M ₁ and M ₂ , and the positional deviation between the substrates 1 and 2 is detected based on the image. If there is a positional deviation, the TFT substrate 1 side is aligned, and the spherical seat 24 is installed on the XYθ table 29 for this purpose.

[0023] Depending on the TV camera 30 in order to clearly photograph the alignment marks M _1, M _2, are configured to illuminate towards the site of these alignment marks M _1, M ₂ in the direction of the television camera 30. For this purpose, two sets of lighting means are provided. The illuminating means includes a light source lamp 31 and a condensing member 32 that condenses the illumination light from the light source lamp 31, and the condensing position of the condensing member 32 is a flexible thin light beam. The incident end 33a of the light guide member 33 in which the fibers are bundled faces. The emission end 33b of the light guide member 33 is arranged opposite to the side surface of the suction plate 22.

In order to illuminate the alignment marks M ₁ and M ₂ , it is necessary to bend the optical path of the illumination light from the exit end 33b of the light guide member 33 by 90 °, and therefore the suction plate 22 has a large number of bends. A flexible light guide 34 is provided, which is a fiber bundle formed by bundling small-diameter optical fibers. Then, the incident end 34a of the light guide 34
Faces the emission end 33b of the light guide member 33, and the emission end 34b
Are routed on the surface of the suction plate 22 so as to face the position below the alignment mark M ₁ of the TFT substrate 1. A horizontal insertion passage 22a and a vertical insertion passage 22b are provided for inserting the light guide 34.

The light guide 34 is flexible so as to change the direction of the optical path, and in order to bend the light guide 34 smoothly and smoothly, the light guide 34 coming out of the horizontal insertion passage 22a is inserted horizontally. It is adapted to be inserted into a vertical insertion passage 22b located away from the passage 22a. Further, the light guide member 33 is also a flexible member, and its position can be adjusted so as to accurately face the incident end 34a of the light guide 34.

In the above structure, first, the TFT substrate 1
Is adsorbed on the adsorption plate 22 which constitutes the lower adsorption means 20, and the CF substrate 2 is adsorbed and held on the upper adsorption plate 12 which constitutes the upper adsorption means 10. Then, by supplying pressurized air from the air supply / exhaust hole 27 to the groove 28,
The spherical portion 23 is lifted from the spherical seat 24. here,
In order to hold the TFT substrate 1 by suction, the inside of the negative pressure chamber 25 is made negative by sucking it from the suction nozzle 26, and at the same time, pressurized air is supplied to the air supply / exhaust hole 27 side. Moreover, since the suction nozzle 26 is loosely fitted in the through hole 23a provided in the spherical portion 23, the negative pressure chamber 25 and the air supply / exhaust hole 27 are held in direct communication with each other. However, the copying of the TFT substrate 1 with respect to the CF substrate 2 is performed within an error range of several μm or less. Therefore, the gap between the through hole 23a and the suction nozzle 26 is Since it is only necessary to have a very small interval to enable such fine copying, the air supply / exhaust holes to which pressurized air is supplied through this insertion part by lengthening the length of this insertion part in the axial direction to some extent. A flow path through which air flows from 27 to the negative pressure chamber 25 is not formed. However, if the pressure difference between the two is too large, an air flow path is formed from the air supply / exhaust hole 27 toward the negative pressure chamber 25. Therefore, in this state, the TFT substrate 1 does not move excessively. Apply a weak suction force that can be retained.

In this state, the TFT substrate 1 and the CF substrate 2 are aligned with each other while keeping a predetermined gap.
In order to make this alignment possible, a television camera 30 is provided above the upper suction means 10, and the illumination light from the light source lamp 31 is guided by the light guide member 33.
The light is emitted from below the CF substrate 2 toward the television camera 30 via the light guide 34. This gives C
F so the alignment mark M ₁ of the alignment mark M ₂ and the TFT substrate 1 of the substrate 2 is lifted, the television camera 30, these alignment marks M
₁ and M ₂ are clearly photographed, and the positional deviation is detected by performing image recognition. Then, by driving the XYθ table 29 based on the result of this image recognition, the TFT substrate 1 is displaced so that these alignment marks M ₁ and M ₂ are aligned, and the position adjustment is performed.

When the TFT substrate 1 and the CF substrate 2 are aligned, the upper suction means 10 is lowered to bring the CF substrate 2 into contact with the TFT substrate 1. Here, the suction plate 22 on which the TFT substrate 1 is sucked is connected to the spherical portion 23, and since the spherical portion 23 is floated from the spherical seat 24 by the pressurized air, the CF substrate 2 is the TFT substrate. When bonded to 1, the TFT substrate 1 comes to follow the CF substrate 2. Here, since the suction plate 22 and the suction plate 12 both have almost exactly parallelism,
The copying of the T substrate 1 is extremely small. Therefore, the spherical portion 23 and the suction plate 2 are required to perform the fine copying.
2 must be held in a state where it does not come into contact with any member. Since the spherical portion 23 is floated from the spherical seat 24 by the pressurized air and the suction nozzle 26 is loosely fitted in the through hole 23a of the spherical portion 23, the spherical portion 23 and the suction plate 22 should be completely floated. Can
The TFT substrate 1 can accurately follow the CF substrate 2.

In this way, the alignment marks M ₁ , M ₂
When the image is captured, the illumination light is transmitted from the light guide member 33 to the light guide 34 in a non-contact state, although it is illuminated from below. Therefore, the suction plate 22
Since the spherical portion 23 is not provided with a member connected to the fixed side, the floating function of the suction plate 22 is not impaired.

When the TFT substrate 1 is made to accurately follow the CF substrate 2, the supply / exhaust holes 27 are made to have a negative pressure. As a result, the spherical portion 23 is adsorbed and fixed to the spherical seat 24 while leaving the TFT substrate 1 and the CF substrate 2 in a state of being traced. At this time, the pressure in the negative pressure chamber 25 is further reduced to
The suction force by the negative pressure on the FT substrate 1 is made stronger. In this state, by pressing the CF substrate 2 by the upper adsorption means 10, the sealing material 3 interposed between the TFT substrate 1 and the CF substrate 2 is compressed. here,
Since the TFT substrate 1 and the CF substrate 2 are accurately parallel to each other, the compressive force to the sealing material 3 is equalized, and the gap G regulated by the spacer 4 mixed in the sealing material 3 is accurate. It is stuck in the state of being kept at.

In this state, the sealing material 3 is cured by irradiating the portion to which the sealing material 3 is supplied with ultraviolet rays by an ultraviolet irradiation device (not shown), and the TFT substrate 1
And the CF substrate 2 are bonded and fixed with the gap G maintained. Then, the negative pressure on the lower suction means 20 side is released, and while the suction by the upper suction means 10 is maintained, the upper suction means 10 is raised to raise both substrates 1, 2.
Can be taken out in a fixed state by being attached.

[0032]

Since the present invention is configured as described above, it is possible to incorporate a light source for clearly photographing the alignment mark with the television camera into the floating mechanism without disturbing the floating function. Play.

[Brief description of drawings]

FIG. 1 is an external view showing a TFT substrate as a lower member and a CF substrate as an upper member.

FIG. 2 is a cross-sectional view showing a state in which a TFT substrate and a CF substrate are bonded together.

FIG. 3 is a cross-sectional view of a bonding device that bonds a TFT substrate and a CF substrate.

[Explanation of symbols] 1 TFT substrate 2 CF substrate 3 Sealing material 4 Spacer 10 Upper adsorption means 11 Adsorption hole 12 Adsorption plate 15 Negative pressure chamber 20 Lower adsorption means 21 Adsorption hole 22 Adsorption plate 23 Spherical part 23a Through hole 24 Spherical seat 25 Negative Pressure Chamber 27 Supply / Exhaust Hole 28 Groove 30 TV Camera 31 Light Source Lamp 33 Light Guide Member 34 Light Guide

Claims (3)

[Claims] 1. A lower member is attached to a vacuum suction portion of a floating member, and an upper member sucked by a vacuum suction means is arranged at an upper position of the lower member, and positions are set with reference to alignment marks provided on the respective members. After joining, the two members are bonded to each other and the two members are bonded together. The lower member, the upper member and the vacuum suction means are formed from transparent or translucent members, and the vacuum is applied. A television camera for image recognition is installed above the suction means, and in the vacuum suction portion, the emission end face of the illumination light faces under the portion of the lower member where the alignment mark is located, and the incident end face faces the side face. A laminating apparatus having a structure in which a light guide made of an optical fiber is mounted and a light source is arranged so as to face the incident end face without contact. 2. The floating member includes a spherical portion connected to the vacuum suction portion of the lower member and a spherical seat fitted to the spherical portion and having an air supply / discharge port opened. , Pressurized air is supplied from the air supply / discharge portion to make the spherical portion in a floating state, the upper member and the lower member are aligned with each other, and both members are joined to each other, The bonding apparatus according to claim 1, wherein the supply / discharge unit has a negative pressure, and the two members are bonded together. 3. The light source comprises a light source lamp and a light guide member made of a flexible optical fiber for transmitting illumination light from the light source lamp, and the optical fiber is provided in the vacuum suction section. 2. The laminating apparatus according to claim 1, wherein the laminating device has a structure in which the position can be adjusted so as to face the incident end surface of the light guide.