TW200937006A - Inspection system - Google Patents

Abstract

The present invention provides an inspection system capable of carrying out the check and inspection and realizing the check of time contraction point. The present disclosed inspection system contains a defect inspection apparatus and a check inspection system. The afore-mentioned defect inspection apparatus recognizes the defect formed on the substrate and acquires the defect information. The defect information includes the defect location coordinate denoting the location coordinate of the afore-mentioned defect and the defect size denoting the size of the afore-mentioned defect. Further, the afore-mentioned defect location coordinate obtained by the afore-mentioned check inspection apparatus according to the afore-mentioned defect inspection apparatus makes the photographing scope relatively move and uses the microscope to check the afore-mentioned substrate; moreover, the afore-mentioned check inspection apparatus contains a coordinate reckoning section, which calculates the coordinate so that the times of the afore-mentioned relative movement can be decreased, according to the photographing scope information used to constitute the afore-mentioned defect information and photographing scope.

Description

200937006 VI. INSTRUCTIONS: I: TECHNICAL FIELD OF THE INVENTION FIELD OF THE INVENTION The present invention relates to a defect 5 check and correction device having a check and correction function, which is expanded by a microscope to be fabricated on a semiconductor wafer. Or a fine pattern on the glass substrate, and the correction function can use a laser to correct the defect. [Prior Art 3 ® Background Art 10 Conventionally, a semiconductor 1C wafer or a liquid crystal panel has been manufactured by a plurality of manufacturing steps. There are generally inspection steps between these steps to manage

A defect in the fine pattern produced by the V device or its manufacturing steps. The inspection step is performed by an inspection system including an appearance inspection device, a line width measurement inspection device, and a check inspection device. The visual inspection device 15 checks for irregularities in the wiring pattern or foreign matter, and the width measurement inspection device measures the wiring pattern. The distance or width, and the inspection and inspection device performs a detailed inspection of the defects detected by the visual inspection device. Early detection of defects during these inspection steps and feedback to the manufacturing steps will affect the performance of the production line. Moreover, the mechanism for improving the performance sometimes has a correction step of correcting the defect. In the inspection system without the correction step, the semiconductor 1C wafer or the liquid crystal substrate panel including the detected defect is judged to be defective and cannot be a product. However, by using the correction device, the defect or the defect can be corrected, and the substrate that is judged to be defective in the inspection step can be regarded as a good product and then entered. A check check device having no correction step is disclosed, for example, in the patent document. The system corrects the error of the defect coordinates output by the inspection device and the observation coordinates of the inspection inspection device, and accurately performs the defect exploration for the specific defect portion 5, and then performs the SEM according to the tendency of the detection value of the defect size. The defect of high magnification observation is selected to improve the efficiency of checking. Further, the correction inspection apparatus having the inspection step includes, for example, a defect correction method and a defect correction device disclosed in Patent Document 2. The correction device analyzes the defects detected by the inspection device as a component of a position coordinate, a shape, or a size of the defect, and uses the data, the circuit design configuration data, and each defect in which the defect is stored. The defect correction method knowledge database of the correction method specifically classifies the defects causing the electrical unconformity and automatically removes or repairs the defects. [Patent Document 1] Japanese Laid-Open Patent Publication No. JP-A-2006-145269 (Patent Document 2) Japanese Patent Laid-Open Publication No. Hei. No. 2006-303227 C. The disclosure of the present invention The inspection device is an observation device that can expand the viewing defect, such as an optical microscope, and is used to check the defect, and the inspection device is inspected by the automatic macroscopic inspection device. In addition, when the automatic giant inspection device has a classification function, its classification data is used to pre-screen the defects of the 200937006 core, thereby reducing the number of inspections. There is a problem in checking by the above method, that is, it is necessary to move the aforementioned microscope to cause each defect outputted by the automatic giant inspection device to enter the center of the field of view, and to move the microscope to check the number of defects. In view of the above problems, it is an object of the present invention to provide an inspection system which can effectively perform a check inspection and which can shorten the inspection time. Means for Solving the Problem 〇 In order to solve the above problems, the present invention adopts the following configuration. That is, according to one aspect of the present invention, the inspection system of the present invention is an inspection system having a defect inspection device and a check inspection device. Further, the defect inspection device identifies a defect formed by the substrate and acquires defect information, and the distortion information includes a defect position coordinate indicating a position coordinate of the defect and a defect size indicating a size of the defect. Further, the check inspection device 15 is configured to move the imaging range relative to each other according to the defect position coordinates obtained by the defect inspection device, and to inspect the substrate with a microscope, and the ® i check mark has a coordinate calculation unit, the coordinate The calculation unit calculates the coordinates 'in accordance with the imaging range information for configuring the defect information and the imaging range to reduce the number of relative movements. Further, in the inspection system of the present invention, the coordinate calculation unit is preferably provided by a device different from the inspection inspection device, for example, a defect inspection device for the network and the inspection device, and a production management device. Or coordinate management server. Further, in the inspection system of the present invention, it is preferable that the coordinate calculation unit has a defect extraction unit in the same field of view, and the same-field-of-view defect extraction unit obtains a coordinate in which a plurality of defects occur simultaneously in the same imaging range of the above-mentioned 5 200937006 microscope. Further, in the inspection system of the present invention, the coordinate calculation unit preferably has a lattice coordinate setting unit. The lattice coordinate setting unit is configured to obtain a lattice in which a plurality of defects appear based on the imaging range of the microscope. Further, in the inspection system of the present invention, it is preferable that the inspection and inspection device has a defective coordinate storage portion that memorizes a defect position coordinate outputted by the defect inspection device. Further, in the inspection system of the present invention, it is preferable that the inspection and inspection device has a spatial modulation element as a verification inspection device having a correction function capable of illuminating a plurality of laser beams in an arbitrary shape at a time. Further, in the inspection system of the present invention, the inspection and inspection device preferably has a portion for not detecting the laser portion, and the portion for which the laser portion is not required to be irradiated extracts a defect portion requiring laser repair, and the portion which does not need to be repaired is removed. Effect of the Invention According to the present invention, since the plurality of defect coordinates on the substrate generated by the semiconductor manufacturing step can be integrated into one coordinate in the same ''photographing range' of the microscope entering the checking and checking device, the relative movement for checking the inspection can be reduced. The number of shots taken and the inspection time is shortened. t lean mode 3 20 Field BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In all the drawings, the same or equivalent components are designated by the same reference numerals, and the description of the common parts will be omitted. (First embodiment) 200937006 Fig. 1 shows a configuration in which an inspection system of the present invention is applied. In the first drawing, the inspection system is a part of a liquid crystal substrate production system for producing a liquid crystal substrate, and has a defect inspection device 101 and a check inspection device 103 connected to the network 104. Further, the network 104 is connected to the production material management server 102 and the coordinate management server 105. Ο 10 The defect inspection device 1〇1 is an automatic giant inspection device, which is characterized in that a linear sensor is used to detect a defect on a substrate and obtain defect information, and the defect includes a defect indicating a position of the defect. The position coordinates and the size of the defect that does not describe the size of the defect. The inspection and inspection device 1〇3 carries a microscopic inspection device that can observe defects at a high magnification, such as a microscope, and observes and inspects the substrate by relatively moving the imaging range with the microscope based on the coordinates of the defect position obtained by the defect inspection device 101. 15 ❹ Further, the production data management feeder 102 has a database function for managing the production line information of the liquid crystal substrate as a whole, and the coordinate management vessel 105 has a database function for managing the verification data. In general, a liquid crystal display device such as a liquid crystal television is manufactured by forming a thin film layer on a glass substrate and repeating a manufacturing step such as a pattern forming step. And the 'inspection step' is an evaluator who performs the pattern formation. The general inspection process is as follows. 20 First, the patterned glass substrate produced by the photolithography step is transported to the manufacturing factory for inspection of defects (4). Next, the defect inspection device _ «set inspection condition (inspection method) performs a comprehensive inspection of the glass substrate', and the unevenness of the coating, foreign matter mixing, or pattern irregularity is detected as a defect. Moreover, Lai check device 1G1 is closed like FTP (FileT_fer 7 200937006

The protocol is registered in the coordinate management server 105 via the network 104. The coordinate management server 105 converts the result of the check into a format that can be registered in the production material management server 102, and registers it as a file with the FTp to log in to the production material management server 1〇2 via the network 104. Then, the substrate on which the pattern defect has been inspected by the defect inspection device (automatic macro inspection device) 101 as described above is transported to the check inspection device 103 by the transport system. Further, the checking and checking device 103 creates a request file for identifying the substrate ID of the transferred substrate as the search condition, and requests the coordinate management 10 server 105 to check the information by the FTp. The coordinate management server 105 obtains the check file from the production material management server 1〇2 by FTP. Further, the coordinate management server 〇5 creates an optimal check file based on the conversion information set internally, and transmits the check information to the check check device 103. 15 The check inspection device 103 photographs the defective image based on the obtained optimal check file and the microscope head of the coordinate moving microscope. Further, the result of classifying the defect type by the image processing function is made into a check result file, and is registered in the coordinate management server 105. The coordinate management server 105 is converted into a format that can be registered in the production material management server 102, and then registered to the production data management server 102 via the network 1 to 4 by FTP. Next, the substrate is carried out by the inspection and inspection device, and the inspection is ended. The present invention can be implemented in the above inspection system. Here, the first embodiment will be described in outline. 200937006 First, the check device 103 receives the missing position coordinates extracted by the defect inspection device 101. Next, check the money check (8) to enlarge the microscope and take the defect received by the missing inspection device 1 () 1 and image the image after processing. The defect of the laser is required to be laser-red (pure) (like 5 e 10 15 20) The absence of fo) is removed from the correction target illuminated by the laser. A defect that does not require t is a defect that is not short-circuited even if it is not repaired, for example, as will be described later. Then, the position of each defective image detected by the defect inspection device 101 is calculated in the order of inspection by the inspection device 103 having the laser repair function. Specifically, the order is sorted according to the coordinates, and it is judged whether there are multiple defects entering the same shooting range. When entering, the plural defects are grouped, and the central coordinates of the group are determined to determine the check. order. Finally, the microscope of the mobile check device 103 is checked in a determined order, and laser repair is performed as necessary. Figure 2 is a diagram showing an example of the format of the check file. In the second figure, the block 2〇1 displays the header field, and the information related to the substrate to be checked includes the following substrate information. In other words, the substrate information is displayed from the front, and the annotation 201a, the substrate ID 201b for specifying the substrate to be checked, and the number of panels 201c included in the substrate are displayed. Block 202 represents defect information for each panel within the substrate. In other words, the defect information is displayed from the front, and the display panel number 202a, the total number of defects 202b included in the panel area, the first defect information 202c, the second defect information 202d, and the third defect information 202e. The first defect information 202c to the third defect 9 200937006 trap information 202e are composed of an index, an X coordinate, a Y coordinate, a χ size, and a γ size, respectively. For example, in the first defect information 202c, the index is "〇1", the X coordinate is "Π1", the Υ coordinate is "111", the X size is "1", and the γ size is "10". Block 5 203 displays defect information for other panel numbers 2 through 4. The above-mentioned checking file is prepared by loading the defective seat mark to be checked into a list-like file and based on the results of the defect inspection device 101 or other inspection device. Next, a method of creating an optimal check file will be described. Ίο Figure 3 is a flow chart showing the flow of the check file creation process for making the best check file. First, in step S301, 'the defect coordinate data of the check file is sorted and made into a coordinate list' so that the microscope can be moved between the defect points in the shortest distance. The reason is that if the defective coordinate data 15 of the check file is processed only by raising or lowering, it is not always possible to move the microscope at the shortest distance for efficient control. Next, in order to minimize the number of movements of the microscope, the defects in the same imaging range (field of view) at the same time as the microscope magnification to be checked are grouped. For the specific steps, first, in step 20 S302, the check list for managing the new coordinate data to be created is initialized. Next, in step S303, the head coordinates of the coordinate list sorted in step S301 are set as the reference defect coordinates. Next, in step S304, the next defect coordinate in the list is compared with the reference defect coordinate, and in step S305, it is determined whether the defective coordinates enter the same shooting range 200937006. When it is judged that the plural defect enters the same-shooting__(5 ❹ 10 15 ❹ 20 I), in the syllabus, the singer is grouped together with the complex defects in the shooting range, and the towel heart of the group is calculated. practice. Next, in order to set the center and coordinates of the group as new reference defect coordinates, the reference defect setting step of step S303 is performed. 7 ... When the trap does not enter the same-shooting range (step S3G5: N.), in step (b), (4) (when it is judged that there is no simultaneous right-minimum _), the base defect is added to the aforementioned management new Check list of coordinate data, face-to-face defect: The target remains as the baseline defect. Then, in the step money summary = No is the last information of the list, when there is more time in the _ list (step: YeS), in order to set the benchmark defect pure candidate to be the new benchmark , _ step coffee benchmark benchmark defect setting steps. On the other hand, when it is judged that it is the last of the coordinate list (step S308: No), the check file creation processing is ended. Next, the processing of the above-described reference defect coordinate setting step (step S3〇3) to the check list addition step (step S3〇7) will be described in detail. 4, 5, 6, and 7 are for explaining the processors of steps S303 to S307 of Fig. 3 in detail. Figure 4 shows the defect 401 and the rectangle 4〇2 of the size into which the defect entered. In the present embodiment, the rectangle 402 is treated as a defect. Fig. 5 shows a state in which the center coordinate of the defect 503 is superimposed on the central coordinate 502 of the microscope field of view (the shooting range) 501, which is the reference defect coordinate set by the step S303 for the first time in Fig. 3, 2009. Moreover, the defect 5〇4 shows a defect close to the defect 503. Fig. 6 shows the state in which the center coordinates have been calculated in step S306 of Fig. 3. Here, the calculated central coordinate 601 is a central coordinate of a rectangle 602 surrounding the defect 5〇3 and the defect 504 included in the microscope field 5〇1. Moreover, the defect 603 further shows a defect close to the defect 5〇3. FIG. 7 shows a state in which the second center coordinate calculation has been performed in step S306, and in addition to the defects 503 and 504, the center coordinates 7 of the group of the microscopic field 501 in which the defect 6〇3 is included are calculated (U. Here, Including the defect 7〇2, that is, the defect group composed of the 〇10 defect 503, the defect 5〇4, the defect 603, and the defect 702 becomes a size outside the microscope field, and thus the defect 5〇3, the defect 5〇4, and the defect 603 As a group, the center coordinate 701 of the group is added to the check-list. The check list created by the processes is used to make the microscope face, grouping multiple defects 503, 504, 6〇 The center coordinate of 7 is moved to the real 15 and the inspection is performed with a minimum number of movements, and the inspection time can be shortened. The present embodiment is to set the coordinate management server 105 to separately check the Xiao Wei. Handling, but for a system that does not cause problems even if the checking of the check case is not caused by the check, it can also be executed by checking the program inside the device 103. Also, the coordinate management servo can be used. 1〇5 pack This function, and the coordinate management server 105 is made into a converted check file. In addition, although the defect is regarded as a rectangle, the curve can be included as long as it surrounds the defect to be observed. Then, the check check device with the pattern correction function Checking block 12 200937006 200937006 ❹ 10 15 ❹ 20 The calculation method is described. Here, the pattern correction function refers to cutting off defects such as crossing two figures (4) by illuminating the defect of the correction target, for example, The special-35, the bullet towel, has a method of cutting the mirror array into an arbitrary shape pattern. When cutting into an arbitrary shape pattern, it is not necessary to use the above-mentioned defects shown in the embodiment as a reference, as long as Considering the wiring pattern, only the area to be subjected to (4) cutting is calculated as the object coordinate, that is, vp. Fig. 8 shows a state in which the anti-residual case is normally formed, and FIG. 9 shows a state in which a defect is formed on the substrate. _. The _ series, showing the state of the defect field extracted by the defect. / In Fig. 8, the pattern 801, the pattern 802, and the __ ship are normally formed on the substrate. In the figure, except for the three normal anti-cases, such as the figure _1, the figure and the pattern 903 formed on the substrate, there are 904, 905 and 906 defects with these patterns. Here, for example, only The image of the normal anti-case shown in Fig. 8 is used as the reference image and is compared with the image containing the defective state as shown in Fig. 9. The difference is as shown in Fig. 1 (10) in the juice. The part of the defect that does not appear on the normal resist pattern can be extracted from the defect field, the defect field 2, the defect field deletion, the defect nano (8) 4, the defect field Na and the defect field 1006. Next, it is used to identify these defect areas. In the _to-defect area _, there is a flow of short-circuit defects that need to be corrected (ie, defects across the pattern) 13 200937006. Fig. 11 is a diagram showing an example of a circuit design configuration data format, and Fig. 12 is a diagram showing an example of a resist pattern. The start position coordinate 1101 of Fig. 11 indicates the start position coordinate (?, ?) of the resist pattern 5 12? 4a of Fig. 12, and the length 1102 indicates the length 1202 of the resist pattern 1204a of Fig. 12. Further, the width 1103 is the width of the resist pattern 1204a' indicating the length 1203 of the resist pattern 1204a of Fig. 12, and the number 1104 indicates the resist pattern of the same size as the resist pattern 1204a defined by the length 1102 and the width 1103. Number. In the example shown in Fig. 12, three anti-surname patterns such as 10 resist pattern 1204a, resist pattern 1204b, and resist pattern 1204c are displayed. Further, the repetition interval 1105 indicates the interval between the resist pattern i2〇4a and the adjacent resist pattern 1204c. The spacing between the resist pattern 1204b and the adjacent resist pattern 1204c is also the interval defined by the repeat interval 1105. Figure 13 is a flow chart showing the flow of short-circuit defect recognition processing for identifying short-circuit defects. The short-circuit defect recognition processing is the aforementioned processing, i.e., processing for identifying a defect in which there is no short circuit in the drawing, and which has no actual influence even if it is not repaired, and a defect which must be repaired. First, in step S1301, a repair list (Repair List) of the rectangular coordinates of the management correction defect 2 is created and initialized, and in step §13〇2, the repetition number of the resist pattern is defined, for example, as shown in the figure The configuration data format § has been described as Number=3. Then, in step S1303, it is determined whether the number is 1 or less, and the step 5 is repeated according to the number of repetitions; the step after 13〇4, when it is determined that the following is 200937006 (step S1303 ·· Yes), the short defect identification is ended. deal with. 5 10 15 Ο 20 On the other hand, when it is judged that it is not 1 or less (step S13〇3: No), in step S1304, the domain coordinates forbidding the short circuit and the prohibited domain coordinates 1 and X2 are calculated. Here, for example, XI is half the width of the center of the pattern of the pattern of the first figure (PointX). This will become the right edge coordinate of the pattern 801. Further, X2 is added to the center coordinate (PointX) of the pattern 8〇1 by adding a pattern repeat interval (interval), and subtracts half of the width of the pattern 8〇1. This will become the left edge coordinate of the pattern 802. Next, in step S1305, it is judged whether, for example, the defect 1001 to the defect 1006 shown in Fig. 1 is located between XI and χ2 calculated in step S1304. Specifically, the left lower coordinate and the upper right coordinate of the rectangle surrounding the defect 1001 and the like are referred to as (DefectL, DefectB) and (DefectR, DefectT), respectively, and it is determined whether or not XI is DefectL or more, and whether X2 is DefectR or less.

Next, when it is judged that the defect 1001 or the like is a defect located between XI and X2 (step S1305: Yes), in step S1306, rectangular coordinates (Repairer L, Repairer R, Repairer B, and Repairer T) are set by the following formula. RepairerL * XI RepairerR : X2 RepairerB · ' DefectB RepairerT : DefectT Further, in step S1307, the rectangular coordinates set in step S1306 are added to the Repair List. Next, in step S1308, the variable Number is incremented by one to identify the next defect, and the interval between the patterns is added to the PointX, and the flow returns to step S1304. 15 200937006 By the short-circuit defect recognition process, defects 1002 and 1004, for example, the figure ίο can be identified as short-circuit defects to be repaired. Thereby, it is possible to prevent the portion that does not need to be repaired from being irradiated. The μ map shows a picture of the state in which the short-circuit defect has entered the same shooting range. In Fig. 14, the defects 1〇〇2 and 1〇〇4 have been identified by the short-circuit defect recognition processing described in Fig. 13, and the processing described in Figs. 3 to 7 is used to judge whether or not the same shooting is entered. Within the scope. Moreover, if the defect 丨〇〇2 and the ❹10 defect 1004 in the microscope field 14〇1 are cut into the correction function of the arbitrary shape pattern by using the micromirror array control as described above, the defect 1002 and the defect 1 can be simultaneously corrected. 4. Reduce the number of coordinate movements and efficiently perform check check for correction purposes. - According to the first embodiment, when the plurality of defects generated in the semiconductor manufacturing step are integrated into a group that enters the same imaging range of the microscope, information on defects that must be corrected can be added, thereby reducing the correction to The purpose of checking the coordinates of the coordinates of the coordinates is checked, and the inspection time can be shortened. Further, according to the first embodiment, a calculation process of integrating a plurality of defects generated in the semiconductor manufacturing step into one group entering the same imaging range of the microscope can be performed separately from the check inspection, whereby the inspection device is The 20 processing load is removed, and the processing time required for the calculation process can be shortened. (Second Embodiment) A second embodiment to which the present invention is applied is a method in which the rectangular shape of the observation region is uniformly placed on the substrate, and the rectangular center coordinates are used as the observation coordinates. The second embodiment can be realized in an inspection system having the same configuration as that of the inspection system phase 16 200937006 of the third embodiment. Here, the outline of the second embodiment will be described. First, the position check coordinate of the defect extracted by the defect inspection device 1〇1 is received by the check inspection device 103. Next, the inspection and inspection device 103 enlarges the microscope and captures the defect received by the defect inspection device 101, and performs image processing on the image by using the short defect identification processing of Fig. 13 described in the first embodiment. In this way, the part that does not need to be repaired is judged, and the part is excluded from the repaired object. 〇 Next, the order of inspection by the check inspection device 103 having the laser repair function is calculated from the position coordinates of the defective image. Specifically, a plurality of lattices (lattice 5 tigers) which cut the substrate into a lattice shape are prepared in accordance with the field of view of the microscope, and the defects detected by the defect inspection device 101 are assigned to the lattice coordinates. For the defect across the complex lattice, since it will become the information of the center coordinates and size, it is judged which other lattice coordinates 15 to enter and the lattice coordinates of the entry are obtained, and the order of the inspection is determined. φ Finally, the microscope of the inspection and inspection device 103 is moved, the inspection is performed in accordance with the determined order, and the laser repair is performed as needed. Fig. 15 is a view showing an example in which a substrate as a check and correction object is divided into a lattice shape based on a rectangular field of view of a microscope field of view. 20 In Fig. 15, the coordinates of the substrate 1501 are the origin (0, 〇) at the lower left, and the X axis for the horizontal axis and the γ axis for the vertical axis. Further, each of the rectangular fields 1502 divided into a lattice shape is provided with a lattice coordinate 15〇3 indicated by an X-row γ column. Figure 16 shows the figure of the defect address of the check file on the substrate. 0 17 200937006 In Figure 16, the defect 16 (Π, defect 16〇2, defect secret, defect i6〇4, defect 1605 and defect 1606] The defect 16〇2 and the defect 16〇3 enter the same-rectangular field (2, 4). Further, the defect 16〇6 is a defect that spans three rectangular fields (7, 〇(8, 1) (9, 1). Fig. 17 is a view showing a sequence in which the microscope moves only in the rectangular direction of the observation field including the defect. In Fig. 17, the dotted line 1702 connecting the start observation position 17〇1 to the final observation position 1703 is shown, and the observation is included in order. Lattice coordinates (丨, 3), lattice coordinates (2, 4), 〇 10 lattice coordinates (3, 6), lattice coordinates (5, 3), lattice coordinates (7, 1) of the center of the rectangular field as the defect of the observation object ), lattice coordinates (8, 1) and lattice coordinates (9, 1). Figure 18 shows a flow chart for determining the order of movement of the microscope to determine the order of movement of the microscope. The sequence of movement of the microscope determines the defects of the processing file. The information must include a rectangular collar that is the defect of the observation correction object. The field is determined by the movement order of the microscope. First, in step S1801, the substrate 1501 is divided into a lattice shape in a rectangular field 1502 corresponding to the observation field of view shown in Fig. 15, and the lattice coordinates are created in the rectangular field. 15〇3 is a quadratic element number of the element number 20 InsPMaP[maxX][maxY]. Here, maxX is the number required to divide the horizontal axis by the horizontal width of the rectangular field 1502, and maxY is the vertical axis. The number of divisions required for the vertical division of the field 1502. Next, in step S1802, the full element data of the secondary element arrangement is initialized. 18 200937006 Next, in step S1803, the check file is read and the check file is read. The full information (defect coordinate, size X, size Y). Here, in order to facilitate the explanation of the coordinate system of the defect, the lower left side shown in Fig. 15 is used as the origin. Next, in step S1804, the subroutine "secondary element" is executed. List 5 InspMap[maxX] [maxY] data setting processing", which is based on the InspMap[maxX][maxY] setting data of the quadratic element matching the defect coordinates. Figure 19 shows the subroutine "secondary matching" Flowchart of InspMap[maxX] ® [maxY] data setting processing", Fig. 20 is a diagram for explaining the defect definition. In Fig. 20, the X dimension 2003 of the defect 2001 is represented by DefectX, and is represented by DefectY. Y size 2004. Again, RectL_X represents the X coordinate 2005X of the lower left coordinate 2005 of the rectangle 2002 surrounding the defect ''2001

RectB_Y represents the Y coordinate 2005Y, RectR_X represents the X coordinate 2006X of the upper right coordinate 2006 15, and the Y coordinate 2006Y is represented by RectT_Y. A subroutine "Secondary Arrangement InspMap[maxX] [maxY] data setting processing" shown in Fig. 19 is a procedure of checking the defective coordinate number cycle shown in steps S1900 and S1912 while changing the coordinates based on the number of defective coordinates. The steps between the steps (ie, step S1901 to step S1911) are in place. First, in step S1901, the X size DefectX2003, the Y size RefectY2004 of the initial defect of the check file, the X coordinate RectL_X2005X of the lower left coordinate of the rectangle 2002, the Y coordinate RectB_Y2005Y, the X coordinate of the upper right coordinate 2006, the RectR_X2006X, and the Y coordinate 19 200937006 are calculated.

RectT_Y2006Y. Next, in step S1902, RectL_X and RectB_Y are set as initial values in the variables PointX and PointY in the subroutine program, respectively. Next, in step S1903, it is calculated which number of elements of the rectangle 2002 coincides with the number of elements of the secondary five-element arrangement InspMap. The number of elements in the quadratic element arrangement InspMap is represented by InspMap[ElementX][ElementY], the initial value of ElementX is set to InitElementX, and the initial value of ElementY is set to InitElementY, and is calculated by the following formula.

InitElementX=PointX/InspAreaX 10 ElementX=InitElement

InitElementY=PointY/InspArea

Element Y=InitElementY Next, in step S1904, the data of InspMap[ElementX][ElementY] matching the number of elements calculated in step 1903 is incremented by one, and in step S1905, ElementX is incremented by one to obtain one. The grid coordinates on the right side (the positive direction of the X-axis), and the starting coordinate value PointX of the ElementX is obtained. Next, in step S1906, it is judged whether or not the start coordinate value PointX of ElementX calculated in step S1905 exceeds the rectangular coordinate RectR_X of the target defect, that is, whether or not the defect is not continuous. If the starting coordinate value p〇intx 20 exceeds the rectangular coordinate RectR_X ', the defect is not continuous. When it is determined that the start coordinate value PointX does not exceed the rectangular coordinate RectR_X (step S1906: No), the data of InspMap[ElementX][ElementY] is incremented by one in step S1907, and the flow returns to step S1905. On the other hand, if it is judged that the starting coordinate value PointX is a rectangular coordinate 200937006

When RectR_X or less (step S1906: Yes), it is judged that the Elememx* has a defect, and the process proceeds to step S1908. 5 Ο 10 15 ❹ 20 Then, in step S1908, the parameter coordinate number ElementX and the start coordinate value PointX of the X axis are returned to the initial value, and in step §19〇9, ElementY is incremented by 1 to obtain one upper side. (the positive direction of the γ-axis) the lattice coordinates 'and find the starting coordinate value p〇intY of the ElementY. Next, in step S1910, it is judged whether or not the start coordinate value PointY of ElementY calculated in step S1909 exceeds the rectangular coordinate RectR_Y' of the target defect, that is, whether or not the defect is not continuous. If the starting coordinate value P〇intY exceeds the rectangular coordinate RectR_Y ', the defect is not continuous. When it is determined that the start coordinate value PointY does not exceed the rectangular coordinate RectR Y (step S1910: No), the data of InspMap[ElementX][ElementY] is incremented by one in step S1911, and the flow returns to step S1905. On the other hand, when it is judged that the start coordinate value PointY is equal to or smaller than the rectangular coordinate RectR_Y (step S1910: Yes), it is judged that the ElementY has no defect. Then, when the number of cycles of the number of defective coordinates ends, the subroutine "InspMap[maxX][maxY] data setting processing" is ended. As described above, the subroutine "Secondary Arrangement InspMap[maxX] [maxY] data setting processing" of the step S1804 of Fig. 18 has been described using the 19th and 20th drawings, and the execution of the data setting processing is performed by the InspMap of the inspection object. [ElementX] [ElementY] No data of 0 or more is entered, so it is only necessary to move the microscope in order to check the center coordinates of InspMap[ElementX] [ElementY] with data other than 0. 21 200937006 Figure 21 shows an example of the rule of microscope movement on a substrate. In Fig. 21, an arrow 2102 on the substrate 2101 shows the movement rule of the microscope. Step S1805 of Fig. 18 is based on the movement rule. Go back to the description of Figure 18. 5 In step S1805, inspMap[ElementX] [ElementY]

ElementX is initialized to 〇. Next, in step S1806, it is determined that ElementX is an even number or an odd number. When it is judged that ElementX is an even number (step S1806: Yes), ElementY is initialized to 〇 in step S1807, and when it is judged that ElementY is an odd number 10 (step S1806: No), ElementY is set to MaxY in step S1808. Next, in step S1809, it is judged whether or not InspMap[ElementX][ElementY] has data entry (1 or more). When it is judged that the data has entered (step S1809: Yes), the center coordinates of InspMap[ElementX] [ElementY] are calculated by the following equation in step S1810 15 as the coordinates of the microscope movement (MoveX, MoveY). MoveX=ElementXxInspAreaX+ InspAreaX/2 MoveY=ElementYxInspAreaY+ InspAreaY/2 Next, in step S1811, the coordinates 2〇 (MoveX, MoveY) calculated in step S1810 are used to create a list for moving the microscope. Then, 'in step S1812', it is judged again that EienientX is an even number or an odd number.

When it is judged that ElementX is an even number (step S1812: Yes), ElementY is incremented by 1 ' in step S1813 and it is judged in step S1814 whether or not ElementY 200937006 is MaxY or more. When it is judged that MaxY is not satisfied (step S1814: No), the process returns to step S1809, and if it is determined to be MaxY or more (step S1814: Yes), the process proceeds to step S1817. 5 ❹ 10 15 ❹ 20 On the other hand, when it is judged at step S1812 that ElementY is an odd number (step S1812: No), ElementY is decremented by 1 in step S1815, and it is judged in step S1816 whether or not ElementY is not full. When it is judged that it is not too full (step S1816: No), the process returns to step S1809, and if the determination is not full (step S1816: Yes), the process proceeds to step S1817. Next, in step S1817, ElementX is incremented by one. Finally, 'in step S1818, 'determine whether or not ElementX is MaxX or higher'. When it is judged that it is not MaxX or more (step S1818: No), the process returns to step S1807', and if it is determined to be MaxX or more (step S1818: Yes), the microscope is terminated. The order of movement determines the processing. According to the second embodiment, since the defect that can be observed at one time is obtained by using the rectangular shape of the observation field created in advance, the load to be calculated can be reduced, and the order of observation can be quickly obtained. In the above, each embodiment of the present invention has been described with reference to the drawings. However, the defect inspection device and the inspection and inspection device provided in the inspection line to which the present invention is applied are not limited to the above embodiments, etc. When it is, it can be a separate unit, a system consisting of multiple devices, or a system that integrates H material LAN and WA. Moreover, a system composed of a CPU connected to the busbar, a memory such as a R〇M or a RAM, a transmission and output device, an external recording device, a media drive device, a portable recording device, and a network connection device can be used. to realise. 23 200937006 In other words, 'a memory such as a ROM or a RAM, an external recording device, or a portable recording medium to which the code of (4) (the system for realizing the foregoing various embodiments (4) is recorded) is supplied to the defect inspection device or the check inspection device, and It is of course also possible to read and execute the code from the computer of the defect inspection device or the check inspection device. At this time, the novel function of the present invention is realized by the code itself read from the configurable recording medium or the like, and the portable recording medium or the like on which the code is recorded constitutes the present invention. The portable recording medium for supplying the code can use, for example, a floppy disk, a ❹ 10 hard disk, a compact disk, an optical disk 'CD-R0M, a CD-R, a DVD-R〇M, a DVD-RAM, a tape, a non-compliant Electrical memory card, R〇M card, various records recorded by network connection devices such as e-mail or PC communication - media, etc. In addition, in addition to the computer to execute the code that has been read into the memory, the function of each embodiment is described in detail, and the actual processing is performed on the computer according to the instruction of the code. The functions of the foregoing embodiments can also be realized by this or all of the processes. ❹ In addition, the program code (data) provided by the provider of the code or program (data) read from the portable recording medium is written into the function expansion unit that has been inserted into the computer or the function expansion unit connected to the computer. After the memory is provided, part or all of the actual processing is performed by the CPU of the function expansion board or the function expansion unit according to the instruction of the code, and the functions of the foregoing embodiments can be realized by the processing. The present invention is not limited to the above-described embodiments 24 200937006 and the like, and various configurations or shapes can be employed without departing from the gist of the invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a view showing the configuration of an inspection system to which the present invention is applied. Figure 2 is a diagram showing an example of the format of the check file. 5 Figure 3 is a flow chart showing the process of creating a check file for the best check file. Fig. 4 is a view (one of the drawings) for explaining the processing of steps S303 to S307 of Fig. 3 in detail. ❹ Fig. 5 is a diagram (the second) for explaining the processing of steps S303 to S307 of Fig. 3 in detail. Fig. 6 is a view for explaining the processing of steps S303 to S307 of Fig. 3 (the third). " Fig. 7 is a diagram (fourth) for explaining the processing of steps S303 to S307 of Fig. 3 in detail. 15 Fig. 8 is a view showing a state in which a resist pattern is normally formed. Fig. 9 is a view showing a state in which a defect is formed on a substrate. Figure 10 is a diagram showing the state of the defect area extracted by the defect. Figure 11 is a diagram showing an example of a circuit design configuration data format. Fig. 12 is a view showing an example of a resist pattern. 20 Fig. 13 is a flow chart showing the flow of short-circuit defect recognition processing for identifying short-circuit defects. Figure 14 shows a picture of the short-circuit defect entering the same shooting range. Fig. 15 is a view showing an example in which the substrate to be checked and corrected is divided into a lattice shape according to the rectangular field of view of the microscope field of view. Figure 16 shows a diagram showing the coordinates of the defect address of the checklist on the substrate. Fig. 17 is a view showing the order in which the microscope moves only in the rectangular field 5 containing the defect. Figure 18 is a flow chart showing the process flow for determining the order of movement of the microscope to determine the order of movement of the microscope. Fig. 19 is a flow chart showing the flow of the subroutine "secondary alignment inSpMap[maxX] [maxY] data setting processing". 10 Figure 20 is a diagram illustrating the definition of defects. Figure 21 is a diagram showing an example of a microscope movement rule on a substrate. [Description of main component symbols] 101... defect inspection device 202c... first defect information 102... production material management server 202d.. second defect information 103... check inspection device 202e... third defect information 104 ...network 203...block 105...coordinate management server 401...defect 201...block 402···rectangular 201a...note 501...microscope view 201b...substrate ID 502···center coordinate 201c·. Panel number 503... Defect 202... Block 504... Defect 202a. Panel number 601... Center coordinate 202b... Total number of defects 602... Rectangular 26 200937006 ❹ 603.. . Defect 701... Center coordinates 702.. Defects 801, 802, 803, 901, 902, 903... Patterns 904, 905, 906 · Defects 1001 Ί 002 '1003 '1004 Ί 005 ' 1006... Defective field 1101... Start position coordinates 1102. Length 1103.. Width 1004.. Number 1005... Repeat interval 1201.. Start position coordinate 1202.. Length 1203.. Width 1204a, 1204b, 1204c, 1205... Anti-speech pattern 1205... Repeat interval 1401...microscope field of view 1501..substrate 1502...rectangular field 1503...lattice coordinates 1601, 1602, 1603, 16 04, 1605, 1606.. . Defect 1701.. Start observation position 1702.. .Dash line 1703.. .. last observation position 2001.. .Defect 2002.. .Rectangle 2003.. .X size (DefectX) 2004·· Υ ( (DefectY) 2005... lower left coordinate 2006.. . upper right coordinate 2101.. substrate 2102... microscope movement rule 27

Claims (1)

200937006 VII. Patent application scope: 1. An inspection system, which has a defect inspection device and a check inspection device; the defect inspection device identifies a defect formed by the substrate, and obtains 5 defect information, where the defect information includes a position indicating the defect a defect position coordinate of the coordinate and a defect size indicating a size of the defect; the check inspection device moves the shooting range relatively according to the defect position coordinate obtained by the defect inspection device, and inspects the substrate with a microscope; 10 The inspection device includes a coordinate calculation unit that calculates a coordinate based on the imaging information of the imaging area for constituting the defect information and the imaging range, so that the number of relative movements is reduced by 2. The inspection of the first item of the patent application is as follows. The system wherein the coordinates of the coordinates are calculated by 15 different devices from the check check device. 3. The inspection system of claim 1, wherein the coordinate calculation unit has a defect extraction portion in the same field of view, and the defect extraction portion in the same field of view obtains a coordinate of the plurality of defects occurring in the same imaging range of the microscope. The inspection system of claim 1, wherein the coordinate calculation unit has a lattice coordinate setting unit for determining a lattice in which a plurality of defects appear based on an imaging range of the microscope. 5. The inspection system of claim 1, wherein the inspection and inspection device has a defective coordinate storage portion, and the defective coordinate storage portion memorizes a defect position coordinate outputted by the defect inspection device of the above-mentioned 28 200937006. 6. The inspection system of claim 1, wherein the inspection and inspection device has a spatial modulation component, and is a correction inspection device having a correction function capable of illuminating a plurality of laser beams in an arbitrary shape at a time for repairing. 7. The inspection system of claim 1, wherein the inspection device has a portion that does not need to be irradiated with a laser, and the portion that does not need to be irradiated with the laser extracts a defect portion that needs laser repair, and the defect is not required to be repaired. Part of it. 29