JP2010097109A - Method for inspecting color filter - Google Patents

Abstract

An object of the present invention is to provide a color filter inspection method that automatically measures the amount of overlap of RGB, which are colored pixels of a color filter, to reduce the work load on the inspector and prevent the occurrence of defects due to the RGB position shift.
A method for inspecting misalignment of a colored pixel of a color filter having a black matrix, the combination of facing sides of the same pixel of the colored pixel from a captured image obtained by imaging the color filter, Alternatively, between different pixels within the same color, the degree of overlap between the colored pixel and the black matrix in any one of the combinations of sides corresponding to the combination of sides facing the same pixel is measured, and the position of the colored pixel Color filter inspection method for inspecting misalignment.
[Selection] Figure 9

Description

  The present invention relates to a method for inspecting misregistration of red R / green G / blue B which are colored pixels of a color filter.

  FIG. 1 is a cross-sectional view showing an example of a color filter substrate 71 used in a color liquid crystal display device. A black matrix (hereinafter referred to as BM) 11 and colored pixels (red R, green G, and blue B) are formed on a glass substrate 10. (Hereinafter referred to as RGB pixels) 12, a transparent electrode 13, and a photo spacer / vertical alignment (hereinafter referred to as PS / VA) 14 are sequentially formed.

  BM11 is a matrix having a light-shielding property, determines the position of RGB pixels of the color filter, makes the size uniform, and blocks unwanted light when used in a display device, The display device has a uniform image with no unevenness and a function of improving contrast. The RGB pixels have red R, green G, and blue B filter functions.

  The BM 11 is formed on the glass substrate 10 by, for example, forming a metal thin film on the glass substrate 10, applying a photoresist to the metal thin film, and then exposing, developing, and developing a pattern having a BM shape by a photolithography method. A method of forming by etching, or a black photoresist resin is applied on the glass substrate 10, and a pattern having a BM shape is exposed and developed by a photolithographic method to develop a so-called resin. A method of forming a pattern called BM is used.

  As a method for manufacturing a color filter having the above-described structure used in many liquid crystal display devices, a photolithography method, a printing method, and an ink jet method are known. FIG. 2 is a schematic block diagram showing steps of the photolithography method. is there. The color filter is a step of first forming BM (BM) on a glass substrate, a step of washing the glass substrate, a step of applying and pre-drying a colored photoresist, a pre-baking step of drying and curing the colored photoresist, and a step of exposing. The step of developing, the step of curing the colored photoresist, the step of forming the transparent electrode, and the step of forming PS / VA are performed in this order.

  For example, when the pixels are formed in the order of R pixel, G pixel, and B pixel, the coloring resist in the order of red R, green G, and blue B from the step of cleaning the glass substrate to the step of curing the colored photoresist. Is repeated three times to form R, G, and B pixels. In the exposure step, the R pixel original is aligned for each glass substrate with respect to the BM already formed on the glass substrate, and then exposed. Similarly, when forming the G pixel and the B pixel, the original plate is aligned and exposed for each glass substrate.

   An example of alignment and exposure is shown in FIG. FIG. 3 shows a color filter in which four color filter patterns 41 are arranged (four faces) on a glass substrate 42. Photoresist applied on a glass substrate using a single photomask (not shown) on a large glass substrate on which the entire four surfaces are drawn when creating a color filter having a pattern of four sides formed A method of exposing the entire surface of a pattern by one exposure is called a batch exposure method. This is because, for example, a photoresist applied on a wafer glass substrate in IC manufacturing is exposed many times by changing the position of the wafer substrate using a photomask on which a single IC pattern is drawn. This is a name for the method of exposing the IC in an impositioned state, that is, the step-and-repeat exposure method. In the batch exposure method, exposure is performed when the alignment mark on the BM pattern and the alignment mark on the photomask are moved to the alignment table on which the photomask is placed and the marks are optically matched. Done. At this time, the matching of the two marks is not always accurately performed, and so-called misalignment may occur.

  Furthermore, in such a batch exposure method, as the glass substrate becomes larger, the photomask used also becomes larger. However, the price of this photomask increases dramatically as the size increases. It will be a thing. In view of this, the four surfaces of the color filter pattern as described above are divided into two surfaces, and divided exposure is performed twice using a photomask having the size of the two surfaces.

  FIG. 4 is a diagram for explaining an example of alignment (positioning) and exposure in the case of divided exposure. 4 includes a carry-in conveyor 51, an alignment stage 52, a carry-in / carry-out robot 53, an exposure apparatus main body 54, a carry-out conveyor 55, and the like. Reference numerals 56a, 56b, 56c, and 56d denote glass substrates on the carry-in conveyor 51, the alignment stage 52, the exposure apparatus main body 54, and the carry-out conveyor 55, respectively.

  FIG. 5 is a side view of the exposure apparatus 54 shown in FIG. As shown in FIGS. 4 and 5, the exposure apparatus main body 54 includes an exposure stage 61, and an exposure unit 62 including a light source 63, a photomask 64, and the like. The position (a) indicates the position of the first exposure in the double divided exposure on the exposure stage 61, and the position (b) indicates the position of the second exposure.

  As shown in FIGS. 4 and 5, the glass substrate 56c on the exposure stage 61 is moved by the exposure stage 61 while the exposure unit 62 is fixed, and the first exposure is performed at the position (a). The exposure stage 61 moves and stops at a certain distance, and the second exposure is performed at the position (b). That is, the glass substrate 56 a is moved from the carry-in conveyor 51 onto the alignment stage 52 (56 b), and is aligned on the alignment stage 52. Next, it is carried onto the exposure stage 61 by the carry-in / carry-out robot 53. The glass substrate 56c on the exposure stage 61 is fixed to the exposure stage 61, and after the above-mentioned two-time divided exposure, it is carried out from the exposure stage 61 onto the carry-out conveyor 55 by the same carry-in / out robot 53 (56d). Yes.

  That is, since the glass substrate is aligned on the alignment stage 52 and then transferred onto the exposure stage 61, the positional deviation on the alignment stage 52, the accuracy of the loading / unloading robot 52, the accuracy of the exposure stage of the exposure apparatus 54, and A positional deviation of the glass substrate due to poor adjustment of the exposure stage of the alignment stage 52, the loading / unloading robot 53, and the exposure apparatus 54 occurs.

  In addition, the orientation of a camera (not shown) for reading marks may be gradually tilted during driving and the reading position may be shifted. In such a case, it is necessary to adjust and restore so that alignment is performed correctly.

FIG. 6 is a diagram for explaining the quality of the result of alignment between BM and RGB pixels. Reference numeral 10 denotes a glass substrate, 11 denotes a BM, 12 denotes a colored pixel (for example, an R pixel), and a case where the R pixel 12 is aligned with the BM 11 will be described as an example. The edge portion 20 of the BM 11 and the edge portion 21 of the R pixel 12 are not a vertical plane but a gentle curved shape. The colored pixels 12 and the BM 11 are aligned by adjusting the so-called overlap amount so that their edge portions overlap each other. A color filter in which the overlap amount is not adjusted well has no contrast when used as a display device. FIG. 6A shows a state where the alignment is good. On the other hand, in FIG. 6B, the overlap amount at the left end of the R pixel 12 is small (conversely, the overlap amount at the right end is large), and in FIG. 6C, the overlap between the BM at the left end of the R pixel and the R pixel is overlapped. There is no amount, and the color filter in which the R pixel is aligned with respect to the BM 11 as shown in FIGS. 6B and 6C becomes defective.

  FIG. 7 shows the amount of overlap between the BM 11 and the R pixel 12 in FIGS. 6A, 6B, and 6C and the state of the overlap portion. 7A, FIG. 7B, and FIG. 7C show the overlap amount between the BM 11 and the R pixel 12, and the lower diagrams show the left side overlap of the BM 11 and the R pixel 12, respectively. A state in which the planar state of the lap portion is seen in a transmission image is shown.

  FIG. 7A shows a case where the R pixel 12 is correctly aligned with respect to the BM 11, and the BM portion 31, the R pixel portion 32, and the overlap portion 33 each showing a transparent image with a normal overlap amount are white. There is no part that appears to be missing. On the other hand, in FIG. 7C, since the R pixel portion 32 is displaced so that the glass substrate can be seen, the overlap portion 33 (in the case of FIG. 7C, there is no overlap portion) appears white. Easy to detect by machine or visual inspection. Further, in FIG. 7B, the position of the R pixel 12 is shifted with respect to the BM 11 and the amount of overlap is small, and the overlap portion 33 is not completely blanked out. Since it is difficult to detect and difficult to detect by an inspection machine or visual inspection, it is necessary for an inspector to check using an image enlargement apparatus or the like.

  However, there is a limit to the number of confirmations to perform the confirmation in the manufacturing process, and even if the confirmation of sampling is performed every certain number of sheets, if there is a sudden misalignment during the sampling confirmation, the next sampling confirmation Production will continue without confirmation.

  Furthermore, the work load to be confirmed by an inspector using an image enlarging apparatus or the like is large, and it has been desired to automatically inspect the overlap amount.

There is Patent Document 1 as a general technical background.
JP 2001-255659 A

  The present invention has been made in view of the above-described problems, and can reduce the workload of an inspector by enabling automatic measurement of the amount of overlap of RGB pixels that are colored pixels of a color filter. It is an object of the present invention to provide an RGB pixel misregistration inspection method that prevents occurrence of defects due to misregistration.

  The invention according to claim 1 of the present invention is a method for inspecting a positional deviation defect of a colored pixel of a color filter having a black matrix, wherein the same pixel of the colored pixel is obtained from a captured image obtained by imaging the color filter. Measure the degree of overlap between the colored pixels and the black matrix in one of the combinations of the sides facing each other, or between the other pixels within the same color, the combination of the sides corresponding to the combination of the sides facing the same pixel Thus, there is a color filter inspection method for inspecting a color pixel misalignment defect.

  The invention according to claim 2 of the present invention is characterized in that relative positional information between a part of the black matrix and a partial region of each of the combined sides is registered in advance. The color filter inspection method according to claim 1.

  The invention according to claim 3 of the present invention measures the degree of overlap of the overlapping portion of the colored pixel and the black matrix, and determines whether or not the color pixel is misaligned based on the presence or absence of overlap, and the overlap is recognized. 3. The color filter inspection method according to claim 1, wherein a second inspection is performed to determine whether the color pixel is misaligned according to the degree of overlap.

  According to the color filter inspection apparatus of the present invention, it is possible to automatically detect the RGB pixel misalignment, reduce the work load on the inspector, and adjust the manufacturing apparatus used in the manufacturing process at the time of misalignment detection. It is possible to prevent the positional deviation of RGB pixels from occurring repeatedly.

  The best mode for carrying out the color filter inspection method according to the present invention will be described below with reference to the drawings.

  FIG. 8 is a conceptual diagram of an apparatus used when a color filter is inspected using the color filter inspection method according to the present invention. It has a table 70 on which a color filter substrate 71 is mounted and moves on a surface plate 74, a camera head 72, a microscope head 73 fixed to and integrated with the camera head 72, and the like. An illumination light source (not shown) that illuminates 71 is located below the color filter substrate 71, and the color filter substrate 71 illuminated by the illumination light source is imaged by a camera head 72 that is an imaging means. Yes. The microscope head 73 is used when the degree of overlap between the colored pixels and the black matrix is visually confirmed. In this embodiment, the camera head 72 is fixed and the color filter substrate 71 is moved, but the color filter substrate 71 may be fixed and the camera head 72 may be moved.

  FIG. 9 is a diagram for explaining a method of registering a measurement field of view and a measurement area. From a captured image obtained by imaging a color filter, the overlapping degree of a colored pixel and a black matrix in a combination of facing sides of the same pixel of the colored pixel It is a figure for demonstrating the case where is measured as an example. Reference numeral 11 denotes a BM, 12 denotes an R pixel, 82 denotes a G pixel, 83 denotes a B pixel, 84 denotes a part of the BM 11, 85 denotes a measurement visual field, 86 denotes a measurement area (1), and 87 denotes a measurement area (2). As the part 84 of the BM 11, it is desirable to select a characteristic part from the BM pattern.

  The measurement area (1) 86 and the measurement area (2) 87 are portions where the BM overlaps with the colored pixel in a partial region of each side of the side combined with the facing side of the same pixel of the colored pixel. The measurement visual field 85 indicates a measurement position in the color filter substrate, and coordinates indicating the position of the measurement visual field 85 are registered in advance.

  A part 84 of the BM pattern having a characteristic shape is registered in advance as a target. The measurement area (1) 86 and the measurement area (2) 87 are arranged so that the part 84 of the characteristic BM pattern, the measurement area (1) 86, and the measurement area (2) 87 enter the measurement visual field 85. Relative position information with respect to a part 84 of the BM pattern having a certain shape is registered in advance.

  The measurement area (1) 86 and the measurement area (2) 87 are registered so as to be at both ends of the R pixel 12 overlapping with the BM 11 respectively.

  Similarly, for the G pixel 82 and the B pixel 83, a part of a BM pattern having a characteristic shape is registered as a target, and relative position information between the target, the measurement area (1), and the measurement area (2) is registered. Is registered in advance.

  By registering the coordinates of the measurement visual field 85 of the color filter substrate in advance, a part 84 of the BM pattern having a characteristic shape registered as a target from the measurement visual field 85 is automatically detected by pattern recognition. The measurement area (1) and the measurement area (2) are determined by the detected target.

  FIG. 9 is a diagram for explaining an example in which the degree of overlap between a colored pixel and a BM in a combination of facing sides of the same pixel of the colored pixel is measured. However, as shown in FIG. Between different pixels, the overlapping degree of the colored pixel and the BM in the combination of the sides corresponding to the combination of the facing sides of the same pixel may be measured.

  That is, as shown in FIG. 10, the combination of sides is not limited to the combination of sides 91 and 92 but may be a combination of sides 91 and 96 or a combination of sides 97 and 98.

  Further, a part in contact with four colored pixels as indicated by 99 may be adopted as a part of a BM pattern having a characteristic shape registered as a target.

  Further, the measurement visual field 90 shown in FIG. 10 (the measurement area is the overlapping portion of the colored pixels and BM in the combination of the sides 91 and 92), and the measurement visual field 93 (the measurement area is the colored pixels and BM in the combination of the sides 94 and 95). As shown by the overlapping portion), it is desirable to register two or more locations in the pattern of the color filter exposed by one shot of the exposure apparatus, so that it is easy to inspect the misalignment of the colored pixels due to the twist. Become.

  A method for detecting positional deviation of RGB pixels will be described with reference to the flowcharts of FIGS. 9 and 11.

  First, target registration and measurement area (1) and measurement area (2) registration are performed (S1). This target is for determining the measurement position, and a part of the BM pattern having a characteristic shape is targeted.

  Next, the substrate is carried into the detection device (S2), and the color filter substrate is moved so that the measurement visual field including the registered target falls within the visual field of the camera head 72 (S3).

  After the camera head 72 moves so that the field of view of the camera head enters the measurement field of view 85 in which the coordinates of the color filter substrate are registered in advance, the target 84 registered in advance is detected by pattern recognition, and the detected target is detected. The measurement area (1) 86 and the measurement area (2) 87 are determined from the positional information relative to 84 (S4). Note that detection by pattern recognition of the target 84 and determination of the measurement area (1) 86 and the measurement area (2) 87 from the target 84 can be performed using existing image software.

  Next, the measurement area (1) 86 and the measurement area (2) 87 are imaged (S5). The measurement area (1) 86 and the measurement area (2) 87 are registered, for example, at both ends of the R pixel 12, that is, at an overlap portion between the BM 11 and the R pixel 12. Next, acquisition of projection processing data (S6) and acquisition of differential processing data (S7) are performed, which are obtained in measurement area (1) 86 and measurement area (2) 87 in FIGS. An example of an image is shown.

  In FIG. 12, the overlap portion is good at both the right end and the left end. In FIG. 13, the overlap portion is small at the left end, and conversely, the overlap portion is excessive at the right end. In FIG. 14, there is no overlap at the left end, and in a display assembled with such a color filter substrate with the base of the color filter substrate exposed on the surface, the left end appears white and missing.

  FIGS. 12A, 13A, and 14A are diagrams in which, for example, an R pixel 12 is aligned with the BM 11, and FIGS. 12B, 13B, and 14B are illustrated. Indicates captured images obtained from the measurement area (1) and the measurement area (2), 31L, 32L, and 33L are captured images obtained from the measurement area (1), and 31L is the left end of the R pixel 12, respectively. BM portion, 32L is an R pixel portion (exactly an R pixel portion excluding a portion overlapping with BM), 33L is an overlap portion of BM11 and R pixel 12, and 31R, 32R, and 33R are In the captured images obtained from the measurement area (2), 31R is the BM portion at the right end of the R pixel 12, 32R is the R pixel portion (exactly the R pixel portion excluding the portion overlapping with the BM), and 33R is the BM11. R pixel 12 It is a Barappu part. FIGS. 12C, 13C, and 14C are graphs of the projection processing data, which represent luminance values with respect to the position of the color filter substrate 10, and SL1 detects white spots. The shresh level 1 is shown. 12 (d), 13 (d), and 14 (d) are graphs of differential processing data obtained by further differentiating the projection processing data into a graph. SL2 is a boundary portion of the overlap portion. A shresh level 2 for detecting (for example, portions shown by 34 and 35 in FIG. 12B) is shown, and A and B show the dimensions of the left and right overlap portions.

  A method for obtaining projection processing data from the captured image of the overlap portion will be described with reference to FIG. FIG. 15A shows an overlap portion of the BM 11 and the R pixel 12, and FIG. 15B shows a captured image of the overlap portion. FIG. 15C is a diagram for explaining a method of acquiring projection processing data from the brightness of the captured image of FIG. 15B, and an analog of the brightness of each captured pixel of the captured image of FIG. 15B. The value is converted into a 256-stage digital value, and further projected in the Y direction (in this case, projection refers to integrating the digital value in the Y direction), and the average value is used as projection processing data. FIG. 15D is a graph of the projection processing data.

  After obtaining the differential processing data (S7), it is determined whether or not there is a white spot (S8). The determination of whether or not there is a white spot is a primary inspection. The white spots are the case where there is no overlap at the left end shown in FIG. 14 and the base of the color filter substrate is exposed on the surface, and for each color of red R, green G and blue B for detecting white spots. White spots are detected by the set shresh level 1 (SL1). If a white spot is detected (YES), the inspection apparatus is stopped (S9), and the exposure apparatus is readjusted so as not to cause a similar white spot defect. After the readjustment, the inspection apparatus is restarted. (S10).

  When there is no white spot (NO), that is, when an overlap is recognized, the following secondary inspection is performed. Differential processing data for measuring the size of the overlap portion and the left end overlap portion dimension A and the right end overlap portion from threshold level 2 (SL2) set for each color of red R, green G, and blue B Dimension B is measured (S11), the absolute value of the difference between A and B is compared with the reference value R of the difference prepared in advance (S12), and the absolute value of the difference between A and B is greater than the reference value R. Is larger (NO), the inspection apparatus is stopped (S9). When the absolute value of the difference between A and B is smaller than the reference value R (YES), it is determined whether all the inspections have been completed (S13). It is carried in (S2) and the inspection is continued. When the total number is finished (YES), the inspection is finished (S14).

In the embodiment of the present invention, a case is described in which projection processing data is obtained from a captured image, and further, differential processing data is obtained from the projection processing data to measure the overlapping degree of a portion where the colored pixel and the BM overlap. However, the present invention is not limited to this, and projection processing may not be performed. For example, one of the five digital value data of the luminance value of the image in the Y direction in FIG. Also, all five digital values of the luminance value of the image in the Y direction are differentiated to obtain 5
An average value of the two pieces of differential processing data may be obtained, and the degree of overlap of the portion where the colored pixel and the BM overlap may be measured.

  As described above, according to the color filter inspection method of the present invention, the RGB pixel position shift can be reduced by automatically measuring the overlap amount, whether or not the overlap amount exists, and the size of the overlap amount. It is possible to prevent the occurrence of defects due to misalignment of the RGB pixels, and when defective products are detected, readjustment of the exposure apparatus in the manufacturing process can prevent the occurrence of repeated misalignment of RGB pixels. .

The top view which showed an example of the color filter used for a color liquid crystal display device. The schematic block diagram which shows the process of the photolithographic method. The figure which shows an example which aligns and exposes. The figure for demonstrating an example of alignment in the case of division | segmentation exposure, and exposure. The side view seen from the direction of the arrow 57 of the exposure apparatus 54 shown in FIG. The figure for demonstrating the quality of the result of position alignment of BM and RGB pixel. The figure which showed the overlap amount of BM and R pixel, and the mode of the overlap part. 1 is a schematic view of a color filter inspection apparatus according to the present invention. The figure for demonstrating the measurement visual field and measurement area concerning this invention. The figure for demonstrating another measurement visual field and measurement area concerning this invention. The flowchart figure for demonstrating the coloring pixel position shift detection method concerning this invention. The figure which shows the case where the amount of overlap is favorable in an example of the image obtained by measurement area (1) and measurement area (2) 87 concerning this invention. The figure which shows the case where there is little overlap amount in an example of the image obtained by the measurement area (1) and measurement area (2) 87 concerning this invention. The figure which shows the case where there is no overlap amount in an example of the image obtained in the measurement area (1) and the measurement area (2) 87 concerning this invention, and a white spot defect. The figure which shows the method of obtaining projection process data from the captured image of the overlap part concerning this invention.

Explanation of symbols

10 ... Color filter substrate 11 ... BM
12 ... RGB pixel 13 ... transparent electrode 14 ... photo spacer (PS) / vertical alignment (VA)
20 ... BM edge portion 21 ... R pixel edge portion 31 ... overlapped BM portion 32 ... overlapped R pixel portion 33 ... overlap portion 31L ... R pixel 12 The leftmost BM part 32L ... R pixel part 33L ... BM11 and the R pixel 12 overlap part 31R ... R pixel 12 rightmost BM part 32R ... R pixel part 33R ... BM11 Overlapping part 34 of the R pixel 12 ... boundary part 35 of the overlapping part ... boundary part 41 of the overlapping part ... pattern 42 of the color filter ... glass substrate 51 ... carry-in conveyor 52 ... Alignment stage 53 ... Loading / unloading robot 54 ... Exposure apparatus body 55 ... Unloading conveyor 56a ... Glass substrate 56b on loading / unloading conveyor 51 ... Glass substrate 56c on Lee placement stage ... the exposure device glass substrate 57 ... arrow 57 on the glass substrate 56d ··· discharge conveyor on the body
61 ... Exposure stage 62 ... Exposure unit 63 ... Light source 64 ... Photomask 70 ... Table 71 ... Color filter substrate 72 ... Camera head 73 ... Microscope head 82 ... G pixel 83... B pixel 84... Part of BM 11 having a characteristic shape 85 .. measurement visual field 86... Measurement area (1)
87 ... Measurement area (2)
90... Measurement field of view 91, 92, 94, 95, 96, 97, 98 .. part of the colored pixel that overlaps BM 93... Measurement area 99.

Claims (3)

  A method for inspecting a color pixel misalignment defect of a color filter having a black matrix, wherein a combination of facing sides of the same pixel of the color pixel or a combination within the same color is obtained from a captured image obtained by imaging the color filter. In other pixels, the degree of overlap between the colored pixels and the black matrix in any one of the combinations of the sides corresponding to the combination of facing sides of the same pixel is measured to check for misalignment of the colored pixels Color filter inspection method.   2. The color filter inspection method according to claim 1, wherein relative position information between a part of the black matrix and a partial region of each of the combined sides is registered in advance. .   The degree of overlap of the portion where the colored pixel and the black matrix overlap is measured, and a primary inspection is performed to determine whether the color pixel is misaligned based on the presence or absence of the overlap. The color filter inspection method according to claim 1, wherein a secondary inspection is performed to determine whether the positional deviation is good or bad.

Images