JP2003520995A - Structure and method for suppressing pixel contamination near seals in AMLCD tiles - Google Patents

Abstract

(57) Abstract In an AMLCD display tile, a polyimide alignment layer determines the stable polarization state of a liquid crystal under zero electric field conditions. If this alignment layer at the pixel is contaminated at or near the seal front by the low molecular weight components of the epoxy seal, a visible defect will occur in the AMLCD display. In a tiled display, the seal is placed as close as possible to the pixel (203) without contaminating the pixel. A unique method of controlling the seal and its formation to reduce or eliminate the aforementioned contamination (202) is the subject of the present invention. Controlling the position of the seal inner circumference (207) to apply a narrow seal to the AMLCD tile provides a high quality seal front.

Description

Detailed Description of the Invention

RELATED PATENT APPLICATION This application is a continuation application of US Provisional Application No. 60 / 177,448. This application was filed in 1999
Co-pending U.S. application No. 09 / 369,465 filed on June 6, 1999 co-pending application 09 / 368,921 filed on August 6, 1999 and co-pending application No. 15 filed on September 15, 1999 60 / 153,962, as well as application 08 / 949,357 filed October 14, 1997 and currently issued as US Pat. No. 5,963,281. FIELD OF THE INVENTION This invention relates to active matrix liquid crystal displays (AMLCD) for assembly into tiled flat panel displays having visually imperceptible seams.
It relates to design structures, materials and methods for manufacturing tiles, and more specifically AM to ensure that the pixels adjacent to the seams are minimally contaminated.
A method for producing a thin inner perimeter seal for LCD tiles. BACKGROUND OF THE INVENTION There are many requirements in the design and processing of tiled flat panel displays that act to have a visually seamless appearance. Optical and mechanical and electrical design parameters are disclosed in detail in US Pat. No. 5,661,531 and co-pending US patent applications 09 / 369,465 and 09 / 368,921. All of these references teach that multiple tiles can be combined to form a large flat panel display with seamless features. In other words, the seams placed between the tiles are visually imperceptible to the viewer.

[0002] One of the most important requirements for a seamless appearance of seams between adjacent tiles concerns the pitch between pixels that enclose the seams. The pitch between pixels on either side of the seam should be essentially equal to the pixel pitch in the central region of the tile. An acceptable level of seam defects for the eye is highly dependent on viewing distance, which is dependent on pixel pitch. Larger pixel pitches and smaller than normal aperture ratios (i.e., emission ratios for dark areas within pixels) give more space to the seam components. Inner circumference of seal (
The SIP) tolerance must take into account the undulation (corrugation) of the seal leading edge near the pixel for two adjacent tiles, as well as other tolerance factors discussed below. The relief is the amount of dimensional change from the straight line of the seal edge.

Seal-edge characteristics depend on 1) seal material, 2) pre-cure conditions, 3) laminating and curing process, 4) panel surface structure, and 5) optimal design of panel surface structure. The seal material is a filled two-part epoxy with glass spacers added for cell gap control and flow control solvent added for flow and cure hardening.

The formation of the seal results in an inner seal edge having three different areas. The volume of seal formed is primarily "filled" epoxy. The edges of the seal include "clear" areas made of low molecular weight unfilled epoxy that are squeezed out during the lamination and curing process. Beyond the front edge of the seal edge, the very low molecular weight bond of residual solvent and epoxy wets the liquid crystal (LC) alignment layer, resulting in LC misalignment. This area is referred to as the "defect area" and can result in undesirable visible artifacts when located near or above the active pixels. Such sealing properties dictate the design constraints for minimizing the LC cell perimeter.

US Pat. No. 5,963,281 and co-pending patent application 09 / 369,465 disclose detailed designs and methods for reducing the relief of the seal leading edge and the associated defect area. This design utilizes dams, channels between dams and wetting structures to help control the formation of the seal leading edge in the SIP area. Nevertheless, this design leaves approximately 50 microns of insurance allocation within each SIP for the defective area. This represents a 100 micron portion of the pixel pitch, which is 10% of the pitch in current 1mm pixel pitch designs.

This reduces the aperture ratio of about 40% pixels in an array of 2 × N tiles,
About 20% reduction in individual tile array. Seams appear to be unidirectional in a 1 × N array. Display brightness depends directly on the pixel aperture ratio, so it is important to maximize the aperture ratio. One way to do this is to reduce the size of the defective area.

As the demand for higher resolution large tiled displays is increasing, the seam area must be reduced to maintain a reasonable aperture ratio. This reduction in seam size necessitates modification of the panel structure near its edges, the epoxy composition and the manufacturing process of the display.

The present invention provides a method for reducing the defective area of the seal inner edge in a SIP, thereby making the finished tile edge seal narrower. This improvement in sticker control provides a 50% aperture ratio in a visually seamless tiled display that can support HDTV (1280 x 720 pixels) and simultaneous XGA (1024 x 768 pixels) in a 1 x N tile array. It can be increased to the front and back. Description of Related Art Several methods are currently used to fabricate seals between thin film transistor (TFT) substrates and color filter (CF) substrates of AMLCD displays. The sealing material, typically a silica-filled epoxy, is mixed with a) a reactive solvent component, b) a volatile solvent component for viscosity control and crosslinking, and c) spacers for cell gap control. The sealing material can be screened on a color filter (or TFT) or dispensed in a desired pattern from a small orifice such as a pen, hollow needle or syringe.

The dispensing program determines the desired pattern, ie the width and height of wetting of the sealing material (ie the volume of the sealing material), on the TFT plate or C
Write on the F board (which is desirable). It is convenient for the TFT plate or CF substrate to have dams and wetting structures placed to help position the seal accurately. Dams and wet structures reduce both the relief and the defect area within the SIP.

Generally, dams have a tolerance (or total corrugation width) range of about 60 microns for a nominal average position of the transparent seal leading edge from pixel to 80 microns for dam dimensions of about 30 microns wide and 1 micron high. Control the front end of the transparent sealing material within. The channel between adjacent dams provides a pass reservoir to smooth the leading edge so that the liquid sealant flows laterally and has a maximum relief of about 60 microns across the tile edge for an 800 micron seal width. Become. In this way, the average relief can be controlled to about 30 microns. Due to such control structure in CF and control of the seal volume and position, the transparent seal material does not flow over the dam close to the pixel. Therefore, contamination and defect kelia are limited to less than the 50 micron guaranteed area between the dam and the pixel, thus preventing contamination by wetting the pixel.

In the dispensing process (as compared to the screening process), variables that affect seal position and volume (ie, volume per unit length of dispensed seal bead) can be more tightly controlled. The following control parameters are available for modern types of dispensing equipment and dispensed seal materials: 1) seal material viscosity, 2) syringe design and diameter, 3) pressure, 4) Z height during dispense. Sensing the height of the syringe on the substrate with feedback to even out, 5) speed control of the dispensing head, and 6) precise alignment of the syringe in the X and Y coordinates.

To set up a target wet seal volume before squeezing or a target seal width after squeezing to a predetermined cell gap between the CF and TFT plates, a wet width break of the bead of seal material after baking is set. It is necessary to measure the area. For current installations, less than 5% seal due to accuracy of wet volume control using these parameters and pressure and / or volume set-up modification using Z-height adjustment to increase volume to target volume Variability in width is obtained.

Since there is some volatility with respect to the reactive solvent mixture, it is desirable to dry before measuring the seal volume. Instead, the elapsed time before measuring the predetermined bead cross section is required to obtain the desired volumetric accuracy. The controlled drying step allows the evaporation of the solvent to a predetermined value. Once setup is complete and the target dispense seal volume has been determined, many panels can be run in a predetermined width with less than 5% variability without further volume measurements and process adjustments.

Pre-baking the seal material at around 90 ° C. for a short time (eg, about 15 minutes) crosslinks the polymer enough to support the assembly of two substrates CF and TFTs ready for lamination. Prior to placement and alignment of the two substrates, the desired spacer spheres with a diameter of about 5 μm create an LC gap and are distributed over the TFT substrate. The sealing material is preferably distributed on the CF substrate. In addition, a second set of spacer spheres or rods (preferably glass) is mixed in the epoxy prior to dispensing. The second set of spacers (preferably spheres) have complementary diameters to minimize variations in the cell gap near the seal.

Next, the TFT substrate is aligned and overlaid on the CF substrate. During this process, the sealing material is extruded from the wet width to the final width when the TFT and CF plates are squeezed or laminated with a selected force or vacuum. For example, a wet width of 200
If the micron and the wet height of the seal material averages 20 microns, then a seal will be 800 microns wide after lamination when a 5 micron spacer is used. In this example, the lamination process moves the inner or leading edge of the seal 300 microns toward the pixel. During lamination, the front end of the seal is undulating (corrugated), and a transparent, unfilled polymeric material appears at the front end of the silica-filled polymeric material. Some wetting of the surface with the reactive solvent causes the appearance of defect areas (not easily detectable) at the front edge of the transparent polymeric material. Therefore, in the defect area, the solvent is thin and only wets the surface and does not bridge the gap between them.

When this contaminant reaches pixels near the seam, visible defects occur. A feature of the defect may be the lack of electronic activity in the contaminated portion of the pixel (ie, the continuously illuminated portion of the subpixel) or a partially illuminated larger area that includes many pixels. is there. Completed with a liquid crystal material filled by placing the AMLCD panel between the uncrossed polarizers (black state) and observing
The defect area on the AMLCD panel can be easily observed. When tested in this condition, the defective area appears to be illuminated.

None of the control parameters described above, except viscosity and solvent content, can be effectively used to reduce the propensity for this type of defect. However, dams and flow directing structures can be employed to limit the defect area to less than 50 microns. In comparison, the conventional method produces a defect area of about 1000 microns.

US Pat. No. 5,963,281 and patent applications 09 / 396,465 and 09 / 368,921 describe dams and () that direct the flow of solvent at the front of the seal on the CF substrate to make the defect area around 50 μm. Or) discloses an example of a channel-like structure. However, in order to narrow the seal width for tiling high resolution AMLCD flat panel displays (FPD), it is desirable to further reduce the defect area to substantially less than 50 μm. The current 50 micron guaranteed area between the dam and the pixel constitutes about 25% of the allowed seam width. Therefore, this area has a large impact on pixel design (eg, reduced aperture ratio, increased pixel size, or reduced resolution) for tiled displays. SUMMARY OF THE INVENTION The present invention teaches a method for reducing the size of the defective area at the edge of a seal, near the edge pixels of a SIP area in an AMLCD tile. This inventive method
Allows the formation of very narrow perimeter seals on individual display tiles designed to be assembled into tiled flat panel displays (FPDs).
The method of the present invention works well with control structures such as dams and wetting structures. Individual hermetically shielded tiles made in accordance with the present invention are suitable for use in tiled flat panel displays having visually imperceptible seams.

Accordingly, it is an object of the present invention to provide a process method that reduces the defect area, thereby allowing a narrower seal to be produced with a controlled SIP.

It is another object of the present invention to provide a sealing material that is optimized by modifying the molecular weight distribution of the reactive solvent material to reduce the proportion of short chain molecules.

It is a further object of the present invention to design a sealing material with a molecular weight distribution that can be easily manufactured by adhesive vendors at a reasonable cost.

In order to allow the selective diffusion of low molecular weight chains in a reactive, volatile solvent, allowing time for the chains to adhere or vaporize to the reactive epoxy terminations prior to panel lamination, the process sequence “ It is an additional object of the invention to modify the "B-stage" temperature profile.

It is a further object of the present invention to reduce the proportion of reactive solvent components in the sealing material.

It is also another object of the present invention to improve seal formation and minimize pixel contamination by reducing the amount of volatile solvent in the seal material.

It is a further object of the present invention to form a narrow (small volume) seal in the range of 500 to 700 μm that exhibits minimal undulations.

It is an additional object of the invention to use glass spheres instead of the glass rods used as spacers in the sealing material.

[0027]   It is also an object of the invention to reduce the SIP size from 185 to 135 μm.

A complete understanding of the present invention may be gained by studying and referring to the accompanying drawings in conjunction with the detailed description set forth below.

For clarity and brevity, similar elements and components are given the same names and numbers throughout the drawings. DESCRIPTION OF THE PREFERRED EMBODIMENTS Generally speaking, the present invention provides a display with a controlled inner seal (SIP) width.
AMLC, which minimizes the defect area at or near the tile seam area
Features an apparatus and method for manufacturing D display tiles.

The AMLCD tile perimeter seal is formed according to the present invention from a seal material having a lower molecular weight component than the materials used in the prior art. In addition, the adhesive mixture has a portion of low viscosity crosslinking solvent. Surface wetting occurs at the leading edge of the seal material as the seal is squeezed during lamination of color filter (CF) and thin film transistor (TFT) substrates. This solvent is designed to assimilate into a cross-linked matrix of epoxy molecules that make up the seal. It is a reactive cross-linking solvent for epoxies that is designed to be proportionately mixed to give a strong cross-linked epoxy matrix after cure. Prior to curing (crosslinking), the bonding of the reactive solvent and the adhesive reduces the viscosity of the seal and provides the wetting ability of the adhesive epoxy seal. Although the structure is designed to control the spread of the material, the leading edge of the material advances unevenly, leaving at least some edge relief. The degree of undulation of the edge increases with the width of the seal, and for seal widths of 500 to 800 microns, the undulation increases by about 15 microns for every 100 microns of seal width.

FIG. 1 is a graph showing seal edge relief as a function of seal width with and without a flow control structure of the seal material such as dams and dam separation channels.

The sealing material generally consists of an epoxy matrix mixed with a solvent in which the reactive component is proportionally mixed with the volatile component to obtain the target viscosity. In addition, the mixture usually contains a substantial amount of particulate silica and glass spacer spheres or rods.

The sealing material can be applied by screening or by dispensing in a desired pattern from a small orifice such as a pen, hollow needle or syringe. Although it is possible to adhere to the TFT substrate alone or in addition to the CF substrate to the TFT substrate, it is preferable that the seal material is attached to the CF substrate. After application, the seal is prebaked and dried to the target polymerization level, commonly referred to as the "B" stage.

Referring now to FIGS. 2a and 2b, there are shown a top view and a cross-sectional view, respectively, of a region of the epoxy seal near the pixel at the edge of the AMLCD panel. Stacking (ie combine CF with TFT and compress this "sandwich")
And after curing, the seal edge has three different components, namely the center 9 of the seal width.
A silica-filled matrix 200 that makes up more than 0%, a transparent polymer 201 that makes up a small width ratio outside the filled matrix 200, and an AMLCD panel filled with liquid crystals (not shown), crossed polarizers. It appears to be composed of a leading edge (defect area) 202 of contamination that is located between and cannot be easily seen until seen in transmitted light.

Observing the defect area 202 is clearly visible as a color variation in the electrically activated pixel, so that the pixel is visible in the transparent unfilled polymer 201.
You can see that it is contaminated at the front end of. The defective area 202 is formed by wetting and filling the LC alignment microgrooves (not shown) with the contaminated adhesive solvent mixture in the polyimide layer (not shown) of the subpixel 203. This is an area that disturbs the polarization alignment. Random polarization orientations, such as those occurring in the defect area 202, visually produce the effect of the non-active portion of the pixel. For example, in an activated normal white LCD panel, the pixel or subpixel or portion of subpixel 203 having defective contamination 202 always appears illuminated. In an activated normal black LCD panel, the defective area always does not appear to be illuminated.

The filled adhesive material 200 and the transparent adhesive material 201 fill the gap 210 between the CF substrate 204 and the TFT 205 substrate, while the low molecular weight contamination mixture 202 has an L of about 5 microns.
Wetting the surface without sufficient surface tension to bridge the C cell height gap 210.
This wets the microgrooves in the polyimide alignment layer, which forms a very thin coating on the ITO thin film layers (not shown) on both the CF substrate 204 and the TFT substrate 205.

Referring also to FIG. 3, for a typical seal, the size of this contaminated area 202 is about 50 microns beyond the leading edge of the visible polymer seal or the inner edge of the seal. Occasionally, contaminants are free to flow along local features, such as the interconnect lines 206 of the TFT substrate 205, and sometimes undesired visual effects or defect areas of hundreds of microns or more near the entire pixel and the inner edge of the seal. May occur.

Therefore, in prior art seals, it was necessary to place the seal away from the pixel (eg, 1 mm or more). This is a common way to make a laptop display.

A commonly used process for laminating and curing seals is to first pre-bake the seal around 90 ° C. (partial curing or B-stage). This causes the reactive solvent component to partially dry and partially crosslink the epoxy molecules. The seal is now tacky and easier to handle. The substrates are then aligned with one another using fiducial marks for alignment. Since pre-baking increases the viscosity of the seal, the subsequent laminating step and flow of seal material is an extrusion-like step, requiring a combination of pressure and temperature to complete the sealing step.

During the stacking process, the cell gap between the CF plate and the TFT plate closes to the diameter of the spacer without splashes or voids. 2
It is important that the substrates are flat, parallel and provide a consistent cell gap in the seal area. If this is not the case, the seal extrusion will not flow with even directionality, and the center point of the seal width will deviate from the predetermined target position. In order to obtain parallelism, the CF structure is designed to be built around the tile outside the seal at a height equal to CF in the central area of the tile. Spacers to equalize the cell gaps are used here and inside the tile.

The temperature during the lamination process is raised in an oven configured around 180 degrees, where the curing or crosslinking of the sealing material is completed. Generally, the panel is vacuum annealed before being filled with liquid crystal. If the solvent does not cross-link the epoxy matrix well, residual solvent may spill out during the vacuum annealing process. However, in the present invention, an alternative (composition, structure and method) developed to reduce contamination at the seal front and thereby reduce the defective area,
This step can be omitted.

Although the viscosity decreases exponentially with temperature, the progress of the crosslinking reaction is contrary to this during the curing cycle. The last molecule to crosslink is the largest chain and the smallest chain is epoxy.
It diffuses more rapidly in the matrix, finding reactive epoxy chains for binding. If there are too many solvent molecules or no epoxy bond is found, it will be easily extruded in front of the seal front moving by pressure penetration. That is, the less viscous solvent and the lesser molecular weight portion are preferentially extruded to the seal front. Thus, the defect area at the leading edge of the seal edge is determined by the solvent component that is not effectively crosslinked to the epoxy during the B-stage or temperature ramp to reach the final cure state during the lamination process. Therefore, the contamination front is the shortest penetrating component of the solvent or the chain length with the smallest molecular weight distribution in the solvent.

The results of the experiments show that, in fact, the defect area becomes smaller if a partial cure at a temperature well below the commonly used pre-baking temperature of 90 ° C. is used. Proven steps for binding reactive solvents are added to the steps around 50 ° C. The reason is that the lower molecular weight components of the reactive solvent disperse more rapidly at this lower temperature than the higher molecular weight components. Thus, the low molecular weight component selectively finds and attaches to the reactive epoxy molecular chain ends. This modification to the process also reduces volatile as well as reactive solvents prior to stack formation, thus reducing pressure penetration of low molecular weight components that cause fouling of seal fronts in SIP areas.

The AMLCD tile perimeter seal is made in accordance with the present invention from the same materials used in the prior art. However, the component composition is different and is optimized in the process to reduce the defect area. The two lowest molecular weight solvents that form the seal leading edge must be reduced by pressure infiltration during lamination (extrusion).
Volatile solvents added to the seal material to control viscosity for dispensing must be minimized during formulation or by more thorough drying prior to lamination. The reactive solvent must be precisely balanced against the molecular content of the epoxy substrate so that no residual solvent remains after the final curing reaction. To this end, a temperature hierarchy is used that optimizes the ability of the reactive solvent molecules to be dispersed throughout the underlying epoxy matrix and to find sufficient reaction sites for complete bonding.

The seal components are mixed on a lot-by-lot basis for storage and are used in manufacturing in a variety of ways. Solid filler materials, typically SiO _2, have a variety of particle sizes, surface areas and content configurations that vary from lot to lot and from user to user. Moreover, the polymer components can vary widely with respect to molecular content and molecular reactivity as determined by thermal history. Further complicating is that the lamination process varies with respect to time and temperature, force and seal leading edge excursion, depending on the particular design being manufactured.

Accordingly, the present invention relies on a functional test method for determining defect area characteristics for a particular manufacturing process and design to be performed. The procedure is as follows. 1) Determining the defect area at the seal front as illustrated in the figures and discussed above for a particular material, process and tool for a target design with a predetermined seal width, cell gap and wetting structure, And 2) Minimize the defective area by one or more of the procedures described below.

Several steps can be performed to reduce the defect area to achieve a narrower seal with controlled SIP. First, it is conceivable to modify the molecular weight distribution of the reactive solvent material by reducing the proportion of molecules with short chain length.
In order to improve the molecular weight distribution, a method of centrifuging the material and then transferring it to another container can be used. It is clear that the specifications of the reactive solvent should include a molecular weight distribution with the minimum chain length cut. This is a fix that the seller of the seal adhesive component can implement.

Second, steps in the process sequence are included to allow selective dispersion of low molecular weight chains in the reactive solvent, including a typical 90 ° C. prebake plus temperature steps around 50 ° C. The dynamic temperature structure can be modified. This gives the low molecular weight chains time to attach to the reactive epoxy termination or to evaporate before stacking. Instead, an optimized prebake temperature time structure starting well below 90 ° C. is used to selectively bind the low molecular weight component parts of the reactive solvent and increase the amount of solvent vaporized. be able to.

Yet another control method is to reduce the proportion of reactive solvent components to balance the molecular epoxy content without producing excess solvent.

Yet another control method is to reduce the volatile solvent so that no volatile solvent remains after pre-baking.

As previously mentioned, the complementary reduction of the defective area at the front end of the SIP can be achieved by the dams, dam separations and wetting structures already disclosed in the above-mentioned related patents and patent applications. However, in order to utilize the invention with respect to the sealing process, it is desirable that dams and wetting structures designed and engineered into the CF be placed in optimal locations close to the pixels depending on the size of the defect area. this is,
FIG. 5 disclosed in copending patent application No. 09 / 490,776 filed January 24, 2000
Finished tiles and the resulting SIP have a certain width as shown in eg 135
It is easy to understand if the example is controlled to be μm.

Tiles are roughly sized by creating a 1 × 3 tiled FPD array for HDTV and / or XGA resolutions. In this case, the dam is placed about 25 μm away from the pixel as a buffer zone to prevent contamination from reaching the pixel. This is a 25 micron reduction compared to the dimensions used in SDTV tiled FPDs. HDTV resolution dams are also relatively narrow, about 20 microns, spaced about 25 microns, and selectively wet in parallel with the leading edge of the penetrating low molecular weight solvent.
Form a channel in CF.

In addition to these differences, comparing FIGS. 4 and 5 shows that the average seal width after cutting is reduced by 15% in the HDTV design compared to the SDTV design. This indicates a reduction in width from 90 microns to 78 microns. Although the dam structure is designed to control the spread of the material, the leading edge of the material still advances unevenly, leaving at least some edge relief. The degree of edge relief increases with the width of the seal as explained above.

This inventive process modification further involves making a narrow (small volume) seal on the order of 500 to 700 μm wide to reduce undulations and total solvent retained in the seal reservoir. . By introducing 5 micron diameter glass spheres as spacers in the sealing material rather than glass fibers or glass rods, smaller diameter syringes can be used for dispensing than in conventional methods. Thus, smaller seal widths can be reliably made without the risk of creating voids or narrowing in the seal width due to intermittent partial plugging of the syringe, which is typically associated with small diameter syringes. In this way, the center of the seal material can be moved within the cell gap towards the AMLCD edge pixels.

The strength of the seal between the TFT substrate and the CF substrate that is subjected to shearing or bending stress is determined by the mechanical strength of the seal and its adhesive interface. Strength is 4 tiles
Proportional to the total seal width at all edges. Therefore, the tile mosaic is 1
After cutting a large amount of the seal at the edges that form the two seams, and further reducing the overall seal width to reduce SIP corrugations and reduce the defect area, the seal is better than the strength required for processing during manufacturing. I'm worried about my weakness. Therefore, modifications to the sequential draw forming of the seal are desirable to increase strength by redundancy. Therefore, a plurality of seal widths can be applied to the wide side surface of the seal by sequential and continuous drawing of the seal. This procedure complements the removal of the seal by a predetermined narrow seal side cut. Redundancy can also be used on narrow seal sides to provide a support system during cutting.

In the future, an improved wetting structure with multiple dams and channels designed to produce greater capillary force to control the solvent front as the seal is being formed during lamination is planned. For this type of design and process, the bond strength can be optimized by removing the weak material that interfaces with the seal. In this way, it is possible to selectively choose the edge of the seal to be cut from the material hierarchy. For CF substrates, the choice can be made to one of polyimide, ITO, dielectric layer or substrate glass.

In summary, a method of modifying a process by selecting materials and / or solvent ratios and adjusting the manufacturing process can be presented as the following series of steps.
1) Mix the sealing adhesive to a predetermined composition and add a spherical spacer of the first diameter Step 2) Place two glass substrates that are sequentially coated with ITO thin film and alignment layer Step 3) Two Depositing a seal of a predetermined width around one outer edge of the glass substrate of step 4) prebaking the seal at a predetermined temperature and time structure step 5) to a predetermined cell gap Step 6) Attach a second diameter spherical spacer with a target diameter to the outer surface Step 6) Laminate two glass substrates with a predetermined force and / or vacuum Step 7) With a predetermined temperature / time structure Step 8) Cure the seal Step 8) Fill the glass substrate with LC Step 9) Place the substrate between the crossed polarizers and inspect whether the defect area is the target dimension Step 10) First, If Recessed specification greater than the area have been determined in advance, with a small decrement (
1% of solvent component) Reduce seal volume or solvent ratio, steps (3) to (9)
Repeat this cycle, reducing solvent each time, until the pre-determined specifications are met. Second, if the defects are larger than the pre-determined specifications, the pre-baking time in step (4) is about 30%. Repeat the steps (2) to (9) while increasing and adding the prebaking time in 30% increments in step (4) until the predetermined specifications are met. Third, the defect has a predetermined specification. If greater, add a gel stage step in the range of 50 ° C to 70 ° C equal to the pre-bake in time, and repeat this cycle with increasing gel stage time in equal increments until a predetermined specification is met, step And 11) Steps to standardize the above optimized parameters for future displays such as the assembly of tiled flat panel displays In the miscellaneous manufacturing operations,
Clearly, it is desirable to have as little intrusive changes to the work as possible. The preferred step (10) modifies the solvent cut of the seal material outside the assembly line. This can be replaced by an alternative step, which may have a greater impact on the assembly line.

The present invention is not considered to be limited to the examples selected for disclosure, as other modifications are changes that will vary with the particular operating conditions and environment or design, and the true scope of the invention. It will be apparent to those skilled in the art that it includes changes and modifications that do not depart from it.

While the invention has been described above, what is desired to be protected by a patent is as set forth in the following appended claims.

[Brief description of drawings]

FIG. 1 is a graph showing seal relief as a function of seal width with and without dam and wetting control structures on a color filter.

FIG. 2a is a top view of an exemplary epoxy seal between a CF and TFT substrate near a pixel.

Figure 2b   2b is a cross-sectional view of the epoxy seal shown in FIG. 2a.

FIG. 3 is a typical epoxy between CF and TFT substrates near a pixel showing selective wetting of the epoxy that results in defective areas along the interconnection lines on the TFT substrate and along the edges of the three subpixels. It is a top view of a seal.

FIG. 4 shows a typical epoxy seal between CF and TFT substrates near a pixel in an SDTV FPD showing the relationship between dam and wetting structure incorporated into the CF membrane layer to control the inner circumference of the seal. It is a top view.

FIG. 5: CF substrate near pixel and TF in HDTV FPD showing the relationship between dam and wetting structure incorporated into CF film to control SIP dimensions for alternative HDTV / XGA designs.
FIG. 6A is a top view of an exemplary epoxy seal between T substrates.

─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Skinner, Dean W.             United States, New York 13850,             Vestal, Country Club Road             804 F term (reference) 2H089 HA33 LA07 LA15 LA19 LA42                       LA49 MA04 MA05 NA41 NA42                       NA45 PA16 QA15 RA05 SA17                       TA04 TA09 TA12                 2H092 JA24 NA04 PA02 PA03 PA04                       PA08 QA07 RA10

Claims (44)

[Claims] 1. A method for manufacturing an active matrix liquid crystal display (AMLCD) tile used in a flat panel display, comprising: a) using a predetermined tool according to a predetermined process. Manufacturing a first AMLCD tile having a predetermined material; b) determining and analyzing a defect area in the first AMLCD tile; and c) minimizing the defect area and the AMLCD. Modifying one or more of the materials, the process and the tool, depending on step (b) of the determination and analysis, to optimize tile fabrication; and d) the modified material. Manufacturing a second optimized AMLCD tile using one or more of an improved process and improved tools,
A method comprising: 2. e) Standardizing said improved material, improved process and improved tool for use in manufacturing future AMLCD tiles,
A method for manufacturing an AMLCD tile according to claim 1, comprising: 3. The predetermined material is used to form a seal, the predetermined sealing material comprising: i) a silica filled epoxy; ii) a reactive solvent; iii) viscosity control and crosslinking. A volatile solvent for iv) and iv) a first spacer having a first diameter for cell gap control, the method for manufacturing an AMLCD tile of claim 1. 4. The seal material further comprises: v) a second spacer having a second diameter complementary to the first spacer diameter to minimize cell gap variations near the seal. A method for manufacturing an AMLCD tile according to claim 3. 5. The method for manufacturing an AMLCD tile of claim 4, wherein the first and second spacers include spherical spacers. 6. The AMLCD of claim 4, wherein the second spacer comprises glass.
Method for manufacturing tiles. 7. The method for manufacturing an AMLCD tile of claim 1, including a flow control structure to regulate the amount of the material dispensed along the sealing area of the AMLCD tile. 8. The method for manufacturing an AMLCD tile of claim 7, wherein the flow control structure is selected from the group of dams, channels between dams and wetting structures. 9. The method for manufacturing an AMLCD tile according to claim 8, wherein the flow structure is redesigned and optimized according to step (b) of the determination and analysis. 10. The reactive solvent comprises a predetermined molecular weight, the step further comprising: e) modifying the molecular weight distribution of the reactive solvent to reduce the proportion of short chain length molecules. A method for manufacturing an AMLCD tile according to claim 3 comprising. 11. The predetermined sealing material enables the selective dispersion of low molecular weight chains in the reactive solvent to provide time for the chains to bond to the silica-filled epoxy. The method for manufacturing an AMLCD tile according to claim 3, wherein the seal is cured in a prebaking process according to a predetermined temperature profile for a set cure time. 12. The method for manufacturing an AMLCD tile of claim 11, wherein said modifying step (c) comprises adjusting and optimizing said predetermined temperature profile. 13. The method for making an AMLCD tile of claim 11, wherein said modifying step (c) comprises adjusting and optimizing said predetermined cure time. 14. The AMLCD tile of claim 3, wherein said determining and analyzing step (b) comprises determining whether said defective area is greater than a predetermined specification for it. Way for. 15. The method of manufacturing an AMLCD tile of claim 14, wherein said modifying step (c) comprises gradually and iteratively reducing the volume of said seal if necessary. Method. 16. The AMLCD of claim 14, wherein said modifying step (c) comprises gradually and iteratively reducing the proportion of said reactive solvent, if necessary.
Method for manufacturing tiles. 17. A method for making an AMLCD tile according to claim 13, wherein said modifying step (c) comprises incrementally and iteratively increasing said prebake time if necessary. . 18. The step of determining and analyzing (b) includes determining whether the defective area is greater than a predetermined specification for it, and in terms of time, the pre-baking and 12. A method for making an AMLCD tile according to claim 11, comprising the step of performing an essentially equal but low temperature gel stage. 19. The method for manufacturing an AMLCD tile according to claim 3, wherein the seal is formed by producing a bead of the seal material and its wet width cross section is measured. 20. The seal is applied to the substrate by screening.
A method for manufacturing an AMLCD tile according to claim 3. 21. The method for manufacturing an AMLCD tile of claim 3, wherein the seal is applied to the substrate by dispensing. 22. The method for making an AMLCD tile of claim 21, wherein the sealing material is dispensed by a tool having an orifice. 23. An active matrix liquid crystal display (AMLCD) comprising a plurality of tiles and a seam between them, comprising: a) having a predetermined structure according to a predetermined process and having a predetermined material. Manufacturing a first AMLCD tile, b) determining and analyzing a defect area in the first AMLCD tile,
c) the determining and analyzing step (b) to minimize the defective area.
A) modifying one or more of said materials, said processes and said structures, according to d), and d) one or more of said improved materials, improved processes and improved structures. An active matrix liquid crystal display characterized by being manufactured by a process comprising the steps of: manufacturing a second optimized AMLCD tile. 24. The step of e) standardizing the improved material, improved process and improved structure for use in manufacturing future optimized AMLCD tiles. The active matrix liquid crystal display described in. 25. The predetermined material is used to form a seal, the predetermined sealing material comprising: i) a silica filled epoxy; ii) a reactive solvent; iii) viscosity control and crosslinking. 24. The active matrix liquid crystal display of claim 23, comprising: a volatile solvent for iv) and iv) a first spacer having a first diameter for cell gap control. 26. The seal material comprises: v) a second spacer having a second diameter complementary to the first spacer diameter to minimize cell gap variations near the seal. 26. The active matrix liquid crystal display according to claim 25. 27. The active matrix liquid crystal display according to claim 26, wherein the first and second spacers include spherical spacers. 28. The active matrix liquid crystal display according to claim 26, wherein the second spacer comprises glass. 29. An active matrix liquid crystal display according to claim 23, comprising a flow control structure to regulate the amount of the material dispensed along the sealing area of the AMLCD tile. 30. The active matrix liquid crystal display according to claim 29, wherein the flow control structure is selected from the group of dams, channels between dams and wetting structures. 31. The flow control structure comprises the determining and analyzing step (b).
31. An active matrix liquid crystal display according to claim 30, redesigned and optimized according to 32. The method wherein the reactive solvent comprises a predetermined molecular weight, and e) modifying the molecular weight distribution of the reactive solvent to reduce the proportion of short chain length molecules. The active matrix liquid crystal display according to item 25. 33. The predetermined sealing material enables the selective dispersion of low molecular weight chains in the reactive solvent to provide time for the chains to bond to the silica-filled epoxy. 26. The active matrix liquid crystal display according to claim 25, wherein the seal is partially cured in a pre-bake process according to a predetermined temperature profile for a given time. 34. The active component of claim 33, wherein said step of modifying (c) comprises adjusting and optimizing said predetermined temperature profile.
Matrix type liquid crystal display. 35. An active matrix liquid crystal display according to claim 33, wherein said modifying step (c) comprises adjusting and optimizing said predetermined cure time. 36. The active matrix liquid crystal of claim 25, wherein said determining and analyzing step (b) comprises determining whether said defective area is greater than a predetermined specification for it. display. 37. The step of modifying (c) reduces the volume of the seal stepwise and iteratively, if necessary, thereby centering the center of the seal material.
37. An active matrix liquid crystal display as claimed in claim 36 including the step of moving towards a pixel of the AMLCD. 38. The active matrix liquid crystal display according to claim 36, wherein said improving step (c) comprises a step of gradually and iteratively reducing the proportion of said reactive solvent if necessary. . 39. An active matrix liquid crystal display according to claim 35, wherein said modifying step (c) comprises incrementally and iteratively increasing said pre-bake time if necessary. 40. The step of determining and analyzing (b) includes determining whether the defective area is greater than a predetermined specification for it, and in time the pre-bake 34. An active matrix liquid crystal display according to claim 33 including the step of performing an essentially equal but cool gel stage. 41. The seal is formed by creating a bead of the seal material, the wet width cross section of which is measured after prebaking and the volume of the seal material is pre-determined near a pixel of the AMLCD. 26. The active matrix liquid crystal display according to claim 25, which is adjusted to fill the cell gap up to a predetermined position. 42. The seal is applied to the substrate by screening.
The active matrix liquid crystal display according to claim 25. 43. An active matrix liquid crystal display according to claim 25, wherein the seal is dispensed onto the substrate. 44. An active matrix liquid crystal display according to claim 43, wherein the sealing material is dispensed by a tool having an orifice.