JP2009293957A - Pattern defect inspection apparatus and inspection method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a pattern defect inspection apparatus and an inspection method having enhanced pixel resolution and high defect detection performance, by excluding the noise due to radiation. <P>SOLUTION: The pattern defect inspection apparatus includes a first sensor having a plurality of first pixels, arranged at a first pixel pitch along a first direction for outputting first image data of optical images of a body to be inspected; a second sensor having a plurality of second pixels, provided in parallel with a second direction approximately intersecting with the first direction at right angles, arranged at the first pixel pitch along the first direction, and arranged at positions displaced from the first pixel positions in the first direction by 1/N (N is an integer equal to 1 or greater) of the first pixel pitch for outputting second image data of optical images of the body to be inspected; and a defect detection part for detecting defects of the body to be inspected, on the basis of the first image data and the second image data. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a pattern defect inspection apparatus and inspection method for inspecting a photomask or the like used in a semiconductor device or the like.

  With the miniaturization of semiconductor devices, it is required for pattern defect inspection devices that inspect defects in photomasks used in the manufacture of semiconductor devices to detect fine defects on the pattern with higher resolution.

  In the pattern defect inspection apparatus, an enlarged optical image of a pattern is formed by an optical system on a sensor such as a line sensor or a TDI (Time Delay Integration) sensor, and an electrical image signal obtained by the sensor is processed. We are inspecting. In order to realize high-resolution and high-precision inspection, high-resolution optical images are obtained using short-wavelength illumination light in the ultraviolet region and an optical system with a high NA (numerical aperture), and the pixel resolution of the sensor is reduced. It is necessary to obtain a high-resolution image signal.

  Recently, the pattern line width of a photomask for a semiconductor device has become 200 nm or less, and the size of the order of 10 nm by proximity correction patterns such as OPC (Optical Proximity Correction) and SRAF (Sub-resolution Assist Feature). Or a fine pattern of 100 nm or less may exist. As the pattern on the photomask becomes finer in this way, in the photomask defect inspection, the resolution of the image obtained by the sensor is insufficient along with the resolution of the optical system, and the S / N necessary for detecting the defect is secured. There is a problem that the defect detection capability is insufficient.

  In order to improve the resolution of the optical system, improvements by shortening the inspection wavelength and increasing the NA of the optical system are being studied. On the other hand, in order to improve the resolution of the sensor image, the method of simply increasing the optical magnification is insufficient in the amount of light, cannot ensure a sufficient output level, and the sensor operating speed necessary for image input is also insufficient. .

  Furthermore, as the sensitivity of the sensor increases, the probability of a noise event occurring in the image obtained by the sensor increases due to the effects of radiation and cosmic rays that exist in nature, and noise events that should not be detected by pattern defect inspection. Is detected as a pseudo defect.

As a device for evaluating the drawing performance of a lens used in a camera or the like, there has been proposed a lens performance evaluation device in which two line sensors are shifted by a half of the pixel pitch (Patent Document 1).
JP 2002-48678 A

  An object of the present invention is to provide a pattern defect inspection apparatus and inspection method with high defect detection capability by increasing pixel resolution of a sensor and eliminating noise due to radiation or the like.

  According to one aspect of the present invention, a plurality of first pixels are arranged at a first pixel pitch along a first direction, and an optical image of a device under test is converted into an electrical image signal. A first sensor that outputs first image data; and a second sensor that is provided in parallel in a second direction that is substantially orthogonal to the first direction when viewed from the first sensor, and that is provided along the first direction. Arranged at a first pixel pitch and arranged at a position shifted by 1 / N (N is an integer of 1 or more) of the first pixel pitch from the arrangement position of the first pixels in the first direction. A second sensor that has a plurality of second pixels, converts an optical image of the object to be inspected into an electrical image signal, and outputs second image data; the first image data; And a defect detection unit that detects a defect of the inspection object based on the image data of 2. That pattern defect inspection apparatus is provided.

  According to another aspect of the present invention, first image data is generated by detecting an optical image of an object to be inspected at a plurality of first detection positions arranged at a first pitch along a first direction. And provided in parallel in a second direction substantially orthogonal to the first direction when viewed from the plurality of first detection positions, arranged at the first pitch along the first direction, and The plurality of second detection positions arranged at positions shifted by 1 / N (N is an integer of 1 or more) of the first pitch from the first detection position in the first direction. What is claimed is: 1. A pattern defect inspection method comprising: detecting an optical image; generating second image data; and detecting a defect of the inspection object based on the first image data and the second image data. Provided.

  According to the present invention, a pattern defect inspection apparatus and inspection method with high defect detection capability are provided by increasing the pixel resolution of a sensor and eliminating noise due to radiation or the like.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
Note that, in the present specification and each drawing, the same elements as those described above with reference to the previous drawings are denoted by the same reference numerals, and detailed description thereof is omitted as appropriate.

(First embodiment)
FIG. 1 is a schematic view illustrating the configuration of a pattern defect inspection apparatus according to the first embodiment of the invention.
As shown in FIG. 1, the pattern defect inspection apparatus 10 according to the first embodiment obtains from a sensor 200 that converts an optical image of an inspection object (photomask) 150 into an electrical signal, and the sensor 200. A defect detection unit 400 that detects defects based on the obtained sensor data (image data) is provided.

The sensor 200 has a plurality of first pixels arranged at a first pitch along the first direction, converts the optical image 155 of the device under test 150 into an electrical image signal, and outputs the first image signal. A first sensor 210 that outputs image data 212 is included. The sensor 200 further includes a second sensor 220 that converts the optical image 155 of the device under test 150 into an electrical image signal and outputs second image data 222. The second sensor 220 is provided in parallel in a second direction substantially orthogonal to the first direction when viewed from the first sensor 210, and the first pixel arrangement position in the first pixel arrangement direction. And a plurality of second pixels arranged at the same pitch as the pitch of the first pixels at a position shifted by a distance of 1 / N of the pitch (N is an integer of 1 or more).
Then, the defect detection unit 400 detects a defect of the inspection target 150 based on the first image data 212 and the second image data 222 described above, and outputs a pattern defect inspection result 480.

As shown in FIG. 1, the illumination unit 130 illuminates the pattern surface of the inspection object 150 and forms an image on the sensor surface of the sensor 200 by the imaging optical system 140, thereby inspecting the inspection object 150. The optical image 155 enters the sensor 200.
In the illumination unit 130, for example, a laser light source 110 that generates laser light with an ultraviolet wavelength, and illumination light 122 generated from light emitted from the laser light source 110, and illumination that irradiates the object 150 with the illumination light 122. The optical system 120 can be used.
The imaging optical system 140 forms an image of the emitted light from the inspection object 150 on the sensor 200 (the first sensor 210 and the second sensor 220).
In addition, an XY table 160 that holds the inspection object 150 and changes (scans) the relative position between the inspection object 150 and the illumination light 122 and a laser interferometer 530 that measures the position of the inspection object 150 are provided. it can.

The first image data 212 and the second image data 222 are AD converted by the first AD conversion circuit 170 and the second AD conversion circuit 180, respectively. The first image digital data 213 that is the output of the first AD conversion circuit 170 and the second image digital data 223 that is the output of the second AD conversion circuit 180 are the single event detection circuit 300 and the average. Is input to the conversion circuit 350. The single event data (single event detection flag) 302 that is the output of the single event detection circuit 300 and the averaged data 352 (sensor data 178) that is the output of the averaging circuit 350 are input to the defect detection unit 400. The
In the present specification, the first image data 171 in a broad sense includes the first image data 212 and the first image digital data 213 described above. Similarly, the second image data 222 including the second image data 222 and the second image digital data 223 are referred to as second image data 181 in a broad sense.

  The pattern defect inspection apparatus 10 illustrated in FIG. 1 is an example of a die to die inspection method, and the averaged data 352 is input to the reference data generation circuit 500 (data delay circuit). The reference data generation circuit 500 receives position data 532 from the laser interferometer 530 and generates reference data 522 at a position synchronized with the scan position (inspection position) of the inspection object 150 from the averaged data 352. That is, the reference data generation circuit 500 has a data delay circuit function. Then, reference data 522 that is an output of the reference data generation circuit 500 is input to the defect detection unit 400.

  The defect detection unit 400 is based on the first image data 212 (171) and the second image data 222 (181), that is, the single event data 302, the averaged data 352 (sensor data 178), and the reference data 522. Based on this, a pattern defect of the inspection object 150 is detected.

FIG. 2 is a schematic plan view illustrating the configuration of a sensor used in the pattern defect inspection apparatus according to the first embodiment of the invention.
FIG. 2A illustrates a case where a TDI sensor (storage type sensor) is used as the first sensor 210 and the second sensor 220 of the pattern defect inspection apparatus 10. In the first sensor 210, the TDI direction (delay stage direction) is the Y-axis direction, and the plurality of first pixels 211 are arranged at a predetermined (first) pixel pitch in the X-axis direction (first direction). There are multiple arrays. The second sensor 220 is parallel to the Y-axis direction (second direction) substantially orthogonal to the X-axis direction when viewed from the first sensor 210, and the second sensor 220 also has a TDI direction ( The delay stage direction) is the Y-axis direction, and a plurality of second pixels 221 are arranged in the X-axis direction at a predetermined pixel pitch. The plurality of second pixels 221 are arranged in the arrangement direction of the plurality of first pixels 221, that is, in the X-axis direction, by 1 / N (N is an integer equal to or greater than 1) of the arrangement position of the first pixels 211 and the pixel pitch. ) Are arranged at positions shifted by a distance.

The example shown in FIG. 2A is a case where N is 2. In other words, the plurality of second pixels 221 are arranged at positions shifted from the plurality of first pixels 221 by a distance of half the pixel pitch (0.5 pixel pitch).
Note that the second pixel 221 has a pixel pitch × (M + 1) in a direction (Y-axis direction) substantially orthogonal to the arrangement direction (X-axis direction) of the first pixels 211 with respect to the first pixel 211. / N) at a position shifted by a distance (M is an arbitrary integer). That is, the arrangement positions of the first sensor 210 and the second sensor 220 in the Y-axis direction are shifted by a distance of pixel pitch × (M + 1 / N). Here, M is any technically possible integer. Here, the arbitrary technically possible integer is, for example, a technical possibility in which the first sensor 210 and the second sensor 220 do not overlap with each other and a wiring area for each sensor can be provided. Including sex.

FIG. 2B illustrates a case where a line sensor is used as the first sensor 210 and the second sensor 220 of the pattern defect inspection apparatus 10. In the first sensor 210 (first line sensor 250), a plurality of first pixels 211 (first line sensor pixels 251) are arranged in the X-axis direction at a predetermined pixel pitch. The second sensor 220 (second line sensor 252) is in parallel with the first sensor 210, and the second sensor 220 also includes a plurality of second pixels 221 (second line sensor pixels). 253) are arranged in the X-axis direction at a predetermined pixel pitch. In this case as well, the plurality of second pixels 221 are arranged in the arrangement direction of the plurality of first pixels 221, that is, in the X-axis direction, by 1 / N (N is N They are arranged at positions shifted by a distance of an integer of 1 or more.
The example shown in FIG. 2B is similarly the case where N is 2, and the plurality of second pixels 221 are shifted from the plurality of first pixels 221 by a distance ½ of the pixel pitch. It is arranged. Similarly, the arrangement positions of the first sensor 210 and the second sensor 220 in the Y-axis direction are shifted by a distance of pixel pitch × (M + 1 / N).
2A and 2B exemplify the case where N is 2, but the present invention is not limited to this, and N can be any integer of 1 or more.

  Thus, in the pattern defect inspection apparatus 10 of the present embodiment, the relative position between the two sensors (the first sensor 210 and the second sensor) is the pixel pitch / N in the X-axis direction (lateral), They are arranged in parallel with a pixel pitch × (M + 1 / N) shifted in the Y-axis direction (vertical).

Thus, in the pattern defect inspection apparatus 10 of the present embodiment, by using a plurality of sensors as the sensor 200, a single event detected by only one of the sensors can be detected as will be described later. That is, a single event caused by noise due to radiation or the like can be eliminated. In addition, a plurality of sensors are arranged with a pixel pitch / N shift, and when N is an integer greater than or equal to 2, the effective pixel pitch of the sensor 200 is the pixel pitch / N, so that the pixel resolution of the sensor is Improve N times.
Thereby, the pattern defect inspection apparatus 10 of this embodiment can provide a pattern defect inspection apparatus with high defect detection capability by increasing the pixel resolution of the sensor and eliminating noise due to radiation or the like.

  Hereinafter, a case where the sensor used for the sensor 200 is a TDI sensor and N is 2 will be described as an example.

FIG. 3 is a schematic plan view illustrating the configuration of a sensor used in the pattern defect inspection apparatus according to the first embodiment of the invention.
That is, FIG. 3 is a plan view illustrating the structure of the TDI sensor.
As shown in FIG. 3, the TDI sensor 230 used in the pattern defect inspection apparatus 10 according to the first embodiment of the present invention includes a plurality of light receiving units arranged in the vertical and horizontal directions (Y-axis direction and X-axis direction). 231 (pixels). The number P of the light receiving units 231 in the X-axis direction is, for example, 1024, but is not limited thereto. Then, the vertical transfer register 232 in the accumulation stage direction for transferring each charge photoelectrically converted by the light receiving unit 231 in the accumulation stage direction (Y-axis direction), and the charge from the vertical transfer register in the pixel direction (X-axis) 2 sets of horizontal transfer registers 233 and 234 for transferring in the direction). Further, two sets of sensor output units for reading sensor output signals, that is, sensor output 1 upper output unit 271 that outputs sensor output 1 upward, and sensor output 2 upper output unit that outputs sensor output 2 upward. 272, a sensor output 1 lower output unit 273 that outputs sensor output 1 downward, and a sensor output 2 lower output unit 274 that outputs sensor output 2 downward. In this way, the TDI sensor 230 can be provided with reading units in the upward and downward directions. That is, the upper reading unit is the sensor output 1 upper output unit 271 and the sensor output 2 upper output unit 272, and the lower reading unit is the sensor output 1 lower output unit 273 and the sensor output 2 lower output unit. 274.

FIG. 4 is a schematic plan view illustrating the configuration of a sensor used in the pattern defect inspection apparatus according to the first embodiment of the invention.
As shown in FIG. 4, the pattern defect inspection apparatus 10 according to the first embodiment of the present invention includes a first sensor 210 and a second sensor composed of a TDI sensor 230 having a plurality of output stages illustrated in FIG. 3. The sensor 220 is provided.

  The first sensor 210 includes a plurality of first pixels 211, and also includes a sensor output 11 upper output unit 261 that outputs the sensor output 11 upward, and a sensor output that outputs the sensor output 12 upward. 12 upper output part 262, sensor output 11 lower output part 263 that outputs sensor output 11 downward, and sensor output 12 lower output part 264 that outputs sensor output 12 downward.

  On the other hand, the second sensor 220 has a plurality of second pixels 221, and outputs a sensor output 21 upper output unit 265 that outputs the sensor output 21 upward, and outputs the sensor output 22 upward. Sensor output 22 upper output section 266, sensor output 21 lower output section 267 that outputs sensor output 21 downward, sensor output 22 lower output section 268 that outputs sensor output 22 downward Yes.

  These output units 261 to 268 correspond to the accumulation stage direction (Y-axis direction) of the TDI sensor 230 corresponding to the scan direction of the inspection object 150, and the first image data 171 and the second image data 181 are obtained. It is possible to output from the upper direction or the lower direction.

FIG. 5 is a schematic view illustrating the configuration of a single event detection circuit used in the pattern defect inspection apparatus according to the first embodiment of the invention.
As shown in FIG. 5, the single event detection circuit 300 used in the pattern defect inspection apparatus 10 according to the first embodiment of the present invention processes the first image data 171 and the first peripheral 2 × 2 It has a pixel output unit 320, a first level difference calculation unit 322, a first maximum / minimum level difference calculation unit 324, and a first single event determination circuit 326. Further, the second peripheral 2 × 2 pixel output unit 330, the second level difference calculation unit 332, the second maximum / minimum level difference calculation unit 334, and the second single event that process the second image data 181 A determination circuit 336 is included.

  The single event detection circuit 300 receives the first image data 171 and data obtained by delaying the second image data 181 by the delay circuit 310 by the distance of the pixel pitch × M illustrated in FIG. . Then, it is detected whether or not there is a single event that has not occurred in the second image data 181 in the first image data 171, and a first single event detection flag 328 is output. Similarly, the second image data 181 is detected whether there is a single event that has not occurred in the first image data 171, and the second single event detection flag 338 is output.

FIG. 6 is a schematic view illustrating the operation of the single event detection circuit used in the pattern defect inspection apparatus according to the first embodiment of the invention.
FIG. 6 compares the target pixel 601 of the first image data 171 with the 2 × 2 pixels 602 around the target pixel 601 of the second image data 181, and is generated in the first image data 171. This shows a method of outputting a single event as a first single event detection flag 328. The figure also compares the target pixel 603 of the second image data 181 with the 2 × 2 pixels 604 around the target pixel 603 of the first image data 171, and is generated in the second image data 181. This represents a method of outputting a single event as a second single event detection flag 338.

  As shown in FIG. 6, in order to detect a single event occurring in the first image data 171, first, the target pixel 601 of the first image data 171 and the position of the image of the target pixel 601 are determined. The first difference calculation unit 322 calculates the difference between the output levels of the four pixels of the 2 × 2 pixels 602 around the second image data 181 as the center. Then, the first maximum / minimum level difference calculation unit 324 calculates the first maximum level difference MAX1 and the first minimum level difference MIN1.

Next, the first single event determination circuit 326 determines whether the first maximum level difference MAX1 is larger than a preset first maximum threshold value. Further, it is determined whether the first minimum level difference MIN1 is smaller than a preset first minimum threshold value. And the sign of the difference between the first maximum level difference MAX1 and the first maximum threshold and the difference between the first minimum level difference MIN1 and the first minimum threshold are different, or The same is detected.
Then, by the first logic result circuit 327, the logical sum of the threshold determination result of the first maximum level difference MAX1 and the threshold determination result of the first minimum level difference MIN1, and the result of the polarity inversion Compute the logical product of. As a result, the threshold value determination result of either the first maximum level difference MAX1 or the first minimum level difference MIN1 is valid, and the first maximum level difference MAX1 and the first maximum threshold value When the sign of the difference and the sign of the difference between the first minimum level difference MIN1 and the first minimum threshold value are the same, the first single event detection flag 328 is output.

On the other hand, in order to detect a single event occurring in the second image data 181, first, the first pixel centered on the target pixel 603 of the second image data 181 and the image of the target pixel 603. The second difference calculation unit 332 calculates the difference between the output levels of the four pixels of the peripheral 2 × 2 pixels 604 of the image data 171. Then, the second maximum / minimum calculation unit 334 calculates the second maximum level difference MAX2 and the second minimum level difference MIN2.
Next, the second single event determination circuit 336 determines whether the second maximum level difference MAX2 is larger than a preset second maximum threshold value. In addition, it is determined whether the second minimum level difference MIN2 is smaller than a preset second minimum threshold value. And the sign of the difference between the second maximum level difference MAX2 and the second maximum threshold and the difference between the second minimum level difference MIN2 and the second minimum threshold are different, or The same is detected.

  Then, by the second logic result circuit 337, the logical sum of the threshold value determination result of the second maximum level difference MAX2 and the threshold value determination result of the second minimum level difference MIN2, and the result of polarity inversion Compute the logical product of. As a result, the result of the threshold value determination of either the second maximum level difference MAX2 or the second minimum level difference MIN2 is valid, and the second maximum level difference MAX2 and the second maximum threshold value When the sign of the difference and the sign of the difference between the second minimum level difference MIN2 and the second minimum threshold value are the same, the second single event detection flag 338 is output.

A specific example will be described.
7 to 9 are schematic views illustrating the operation of the single event detection circuit used in the pattern defect inspection apparatus according to the first embodiment of the invention.
7 to 9 exemplify the operation of detecting the first single event occurring in the first image data 171 for the sake of simplicity, the first occurring in the second image data 181 is illustrated. Although the operation for detecting the second single event is not shown, the detection of the second single event can be performed in the same manner as the detection of the first single event.
7 illustrates an operation when there is a single noise event due to radiation or the like, FIG. 8 illustrates an operation when there is a defect to be detected in the pattern of the inspection object 150, and FIG. The operation at the edge portion of the pattern of the inspection object 150 in which the luminance of the image changes is illustrated.

  As shown in FIG. 7, when an optical image of the inspected object 150 is obtained by the first sensor 210, a single event (single noise event) due to radiation or the like occurs, and the first image data 171 includes A noise event unit 620 having a higher brightness than surrounding pixels is generated. On the other hand, this noise event due to radiation or the like does not occur when the second sensor 220 obtains the optical image of the object 150 to be inspected. The brightness of the pixels of the image data 181 is the same as the brightness of surrounding pixels.

  For this reason, the luminance difference due to the noise event differs between the luminance of the noise event unit 620 of the first image data 171 and the luminance of the surrounding 2 × 2 pixels 602 of the second image data 181 corresponding to the position of the noise event unit 620. Will occur. That is, the four pixels of the surrounding 2 × 2 pixels 602 are both relatively lower than the luminance of the noise event unit 620. As a result, the first maximum level difference MAX1 and the first minimum level difference MIN1 calculated by the first level difference calculation unit 322 and the first maximum / minimum level difference calculation unit 324 are the first single event determination. Circuit 326 determines that both are greater than the threshold. As a result, the output of the first logic result circuit 327, that is, the first single event detection flag 328 is “ON”, which is generated in the first image data 171 and is generated in the second image data 181. No single event (first single event) is detected.

On the other hand, as shown in FIG. 8, when there is a defect in the pattern of the inspection object 150, both the first image data 171 and the second image data 181 are brighter than the surrounding due to the defect. A defective portion 630 is generated. As a result, the first maximum level difference MAX1 is determined to be larger than the first maximum threshold value, and the first minimum level difference MIN1 is determined to be smaller than the first minimum threshold value. As a result, the first single event detection flag 328 is set to “OFF”.
In this case, the four pixels of the peripheral 2 × 2 pixels 602 of the second image data 181 may not include the defect portion 630, and the first maximum level difference MAX1 is a threshold value. As described above, the first minimum level difference MIN1 becomes negative from around zero and is equal to or less than the threshold value, so that it is not detected as a single event.

  Further, as shown in FIG. 9, the edge portion 640 of the pattern of the inspection object 150 has a change in luminance as compared with the surrounding pixels. At this time, the first maximum level difference MAX1 between the first image data 171 and the surrounding 2 × 2 pixels 602 of the second image data 181 may be equal to or greater than the first maximum threshold value. However, since the first minimum level difference MIN1 becomes negative from around zero and becomes equal to or smaller than the first minimum threshold, it is not detected as a single event. Therefore, the first single event detection flag 328 is “OFF”.

In this way, the single event detection circuit 300 can distinguish the noise event part 620 from the defect part 630 and the edge part 640, and can detect a single event due to radiation or the like.
A single event occurring in the second image data 181 can also be detected by the same operation as described above.

FIG. 10 is a schematic view illustrating the configuration of a defect detection unit used in the pattern defect inspection apparatus according to the first embodiment of the invention.
As shown in FIG. 10, the defect detection unit 400 used in the pattern defect inspection apparatus 10 according to the first embodiment of the present invention has an average based on the first image data 171 and the second image data 181. The defect data 352 (sensor data 178) is compared with the reference data 522, and the pattern defect of the inspection object 150 is detected. The defect detection unit 400 includes an alignment circuit 410 that aligns the sensor data 178 (averaged data 352) based on the first image data 171 and the second image data 181 and the reference data 522. . Further, a circuit for comparing two sets of image data, such as a level comparison circuit 420 and a differential comparison circuit 430, is provided. The level comparison circuit 420 has a function of obtaining a level difference between the sensor data 178 (for example, averaged data 352) and the reference data 522, and comparing the result with a predetermined threshold value to detect a defective portion. is there.

  As illustrated in FIG. 10, the defect detection unit 400 further includes defect determination circuits 440 and 450 that perform defect determination, and a logical sum circuit 460 that calculates a logical sum thereof. Further, there is a logical product circuit 470 that takes a logical product (AND) of the result of the logical sum circuit 460 and the first and second single event detection flags 328 and 338.

As a result, the logical product of the result of the defect determination circuits 440 and 450 and the first and second single event defect flags 328 and 338 is calculated, and the defect determination can be invalidated in the case of a single event. . That is, when a single event that appears only in one of the first image data 171 and the second image data 181 occurs, the generated single event can be excluded in units of pixels.
Thereby, the pattern defect inspection apparatus 10 according to the present embodiment can provide a pattern defect inspection apparatus with high defect detection capability by eliminating noise due to radiation or the like.

Here, the differential comparison circuit 430 used in the pattern defect inspection apparatus according to the first embodiment of the present invention will be described.
FIG. 11 is a schematic view illustrating the configuration of a differential comparison circuit used in the pattern defect inspection apparatus according to the first embodiment of the invention.
As illustrated in FIG. 11, the differential comparison circuit 430 includes a differentiation circuit 431 that differentiates the sensor data 178 (for example, averaged data 352). Thus, the sensor data 178 is spatially differentiated in a total of four directions, for example, the X direction, the Y direction, the + 45 ° direction, and the −45 ° direction, and is output to the selector 432. The differential comparison circuit 430 further includes an edge direction detection circuit 433 that detects an edge direction from the reference data 522, a peripheral pixel edge direction differentiation circuit 434 that differentiates edge direction data of the peripheral pixels from the reference data 522, and a maximum value detection. A circuit 435 and a comparison operation circuit 436 for comparing and calculating the results of the selector 432 and the maximum value detection circuit 435 are provided.

  Thereby, for example, the differential comparison circuit 430 detects the edge direction from the reference data 522, and the surroundings obtained from the reference data 522 differentiated in the same direction from the absolute value obtained by differentiating the sensor data 178 in the same direction as the edge direction. The maximum value of the absolute value of the differential value of the pixel is subtracted, and the result is compared with a threshold value to detect a defective portion, and differential comparison data 438 is output. Then, the differential comparison data 438 is input to the defect determination circuit 450 illustrated in FIG. 10, and defect determination is performed.

(Comparative example)
Hereinafter, a comparative example will be described.
FIG. 12 is a schematic view illustrating the configuration of a pattern defect inspection apparatus of a comparative example.
As shown in FIG. 12, in the pattern defect inspection apparatus 91 of the comparative example, a sensor 290 that converts an optical image of the inspection target (photomask) 150 into an electrical signal, and sensor data ( A defect detection unit 400 that detects defects based on (image data) is provided.

  In the case of the comparative example, the sensor 290 includes one TDI sensor 230 or one line sensor, and image data obtained by the sensor 290 is AD-converted into sensor data 198. The sensor data 198 is input to the defect detection unit 490 and the reference data generation circuit 500. In addition, since the sensor 290 includes one sensor, the single event detection circuit 300 and the averaging circuit 350 illustrated in FIG. 1 of the first embodiment are not provided.

FIG. 13 is a schematic view illustrating the configuration of a defect detection unit used in the pattern defect inspection apparatus of the comparative example.
As shown in FIG. 13, the defect detection unit 490 used in the pattern defect inspection apparatus 91 of the comparative example includes an alignment circuit 410 that aligns the sensor data 198 and the reference data 522, a level comparison circuit 420, and a differential comparison. A circuit 430 and defect determination circuits 440 and 450; The logical product circuit 470 described in FIG. 10 of the first embodiment is not provided.

  Thus, in the pattern defect inspection apparatus 91 of the comparative example, since the sensor 290 is composed of one sensor, noise caused by radiation or the like cannot be excluded, and the detected defect data should be detected. Therefore, the defect detection ability is inferior. In addition, the pixel resolution of the sensor is determined by the pixel pitch constituting the sensor 290, and the pixel resolution cannot be increased.

  In the pattern defect inspection apparatus 91 of the comparative example, if the inspection object 150 is inspected a plurality of times, it is possible to eliminate noise caused by radiation or the like, but this is not practical because the inspection time becomes longer.

  In contrast, in the pattern defect inspection apparatus 10 according to the first embodiment, by using a plurality of sensors as the sensor 200, a single event can be detected, and a single event caused by noise due to radiation or the like can be eliminated. it can. Further, a plurality of sensors are arranged with a pixel pitch / N shift, and when N is an integer of 2 or more, the pixel pitch of the sensor is substantially increased, and the pixel resolution of the sensor is improved.

  Moreover, in the pattern defect inspection apparatus 10 of this embodiment, since a plurality of sensors are provided in parallel and adjacent to each other, it is easy to compare a plurality of image data obtained by the plurality of sensors. The efficiency of detection is increased. Further, since the displacement of the pixel positions of the plurality of sensors is 1 / N (integer) of the pixel pitch, as will be described later, for example, an operation to obtain by interpolating data between pixels, Arithmetic processing of a plurality of image data by a plurality of sensors is facilitated. When N is an integer of 2 or more, the effective pixel resolution can be improved N times.

  When N is an infinite integer, the arrangement position of the second pixels 221 in the X-axis direction is substantially the same as the arrangement position of the first pixels 211 in the X-axis direction, and there is a difference between the two. However, even in this case, the single event can be eliminated, and the effect of the present embodiment can be exhibited. That is, even when there is no position shift in the X-axis direction, the first pixel 211 and the second pixel are separated by a distance of pixel pitch × M (M is any technically possible integer) in the Y-axis direction. 221 is arranged. For this reason, there is a difference in time for scanning a distance of pixel pitch × M between the time at which the specific portion of the inspection object 150 is imaged by the first pixel 211 and the time at which the second pixel 221 is imaged. Even if a single event occurs, this can be detected efficiently.

  Further, since the positional relationship between the first pixel 211 and the second pixel 221 is a mutual relationship, “first” and “second” can be interchanged throughout this specification.

(Second Embodiment)
FIG. 14 is a schematic plan view illustrating the configuration of a sensor used in the pattern defect inspection apparatus according to the second embodiment of the invention.
As shown in FIG. 14, the pattern defect inspection apparatus according to the second embodiment of the present invention includes a first sensor 210 and a second sensor composed of a TDI sensor 230 having a plurality of output stages illustrated in FIG. 3. A sensor 220 is included. Except for this sensor portion, the pattern defect inspection apparatus 10 according to the first embodiment can be the same as the first embodiment, and a description thereof will be omitted.

  Then, the lower reading unit (sensor output 11 lower output unit 263 and sensor output 12 lower output unit 264) of the first sensor 210 and the upper reading unit (sensor output 21 upper output unit of the second sensor 220). And multiplexers (MUX) 281 and 282 connected to the H.265 and sensor output 22 upper output unit 266). That is, the reading units of the first sensor 210 and the second sensor 220 that are arranged adjacent to each other are shared. The output of each sensor is switched as necessary, and the first image data 212 and the second image data 222 are output from a common signal output pin of the sensor package in which the sensor device is mounted.

FIG. 15 is a schematic plan view illustrating the operation of the sensor used in the pattern defect inspection apparatus according to the second embodiment of the invention.
As shown in FIG. 15A, when the output of each sensor is read from the upward direction (backward direction) 201, the output of the first sensor 210 is the sensor output 11 upper output unit 261 and the sensor output 12 upper side. The output from the output unit 262 and the output from the second sensor 220 are read from the output unit 291 of the MUX 281 and the output unit 292 of the MUX 282.

  Further, as shown in FIG. 15B, when the output of each sensor is read from the lower direction (front side) 202, the output of the first sensor 210 is the output unit 291 of the MUX 281 and the output of the MUX 282. Reading from the output unit 292, the output of the second sensor 220 is read from the lower output unit 267 of the sensor output 21 and the lower output unit 268 of the sensor output 22.

  As described above, in the pattern defect inspection apparatus according to the present embodiment, the multiplexer (MUX) 281 connected to the output stage (reading unit) of the first sensor 210 and the output stage (reading unit) of the second sensor 220. , 282, the circuit configuration is simplified and the circuit noise is reduced. Thereby, it is possible to provide a pattern defect inspection apparatus with higher defect detection capability.

(Third embodiment)
FIG. 16 is a schematic view illustrating the configuration of a pattern defect inspection apparatus according to the third embodiment of the invention.
The pattern defect apparatus according to the third embodiment is a Die to Database inspection method apparatus that performs inspection using reference data electrically generated from design data of an object to be inspected 150.
As shown in FIG. 16, the pattern defect inspection apparatus 30 according to the third embodiment includes a data expansion circuit 520 that expands the design data 510 of the inspected object 150. Then, instead of the sensor data 178 illustrated in FIG. 1, data development data 521 that is an output of the data development circuit 520 is output to the reference data generation circuit 501. Since other parts can be the same as those of the pattern defect inspection apparatus 10 according to the first embodiment illustrated in FIG. 1, description thereof is omitted.

Here, a pattern defect inspection method will be described.
FIG. 17 is a schematic view illustrating the inspection method of the pattern defect inspection apparatus according to the embodiment of the invention.
FIGS. 17A and 17B illustrate configurations of a Die to Die inspection method (DD comparison) and a Die to Database inspection method (DB comparison) used in the embodiment of the present invention, respectively.

As shown in FIG. 17A, in the Die to Die inspection method, an optical image of a mask pattern (a pattern of the inspected object 150) obtained by a sensor and a mask pattern obtained by repeating the same pattern are used. Inspection is performed by comparing the optical image. The pattern defect inspection apparatus 10 illustrated in FIG. 1 employs this method.
On the other hand, as shown in FIG. 17B, in the Die to Database inspection method, an electrical image is obtained from an optical image of a mask pattern obtained by a sensor and mask design data 510, that is, CAD (Computer Aided Design) data. A comparison is made with the reference data generated in (1). The pattern defect inspection apparatus 30 illustrated in FIG. 16 employs this method.

  As described above, the pattern defect inspection apparatus 30 according to the present embodiment adopting the Die to Database inspection method also increases the pixel resolution of the sensor, similarly to the pattern defect inspection apparatus 10 adopting the Die to Die inspection method. By eliminating noise due to radiation or the like, a pattern defect inspection apparatus having a high defect detection capability can be provided.

(Fourth embodiment)
FIG. 18 is a schematic view illustrating the configuration of a pattern defect inspection apparatus according to the fourth embodiment of the invention.
As shown in FIG. 18, the pattern defect inspection apparatus 40 according to the fourth embodiment of the present invention includes an interpolation circuit 360 instead of the averaging circuit 350 illustrated in FIG. 1. In addition, Die to Database inspection method is adopted. Except for the interpolation circuit 360, the description is omitted because it is the same as the pattern defect inspection apparatus according to the third embodiment illustrated in FIG.

FIG. 19 is a schematic view illustrating the configuration of an interpolation circuit used in a pattern defect inspection apparatus according to the fourth embodiment of the invention.
As shown in FIG. 19, the interpolation circuit 360 used in the pattern defect inspection apparatus 40 according to the fourth embodiment of the present invention includes a delay circuit 361 and first XY direction interpolation for interpolating the first image data 171. The circuit 363 includes a second XY direction interpolation circuit 364 that interpolates the second image data 181, and an interpolation data merge circuit 365 that interpolates these data. The first image data 171 is directly input to the first XY direction interpolation circuit 363. The second image data 181 is input to the Y axis of the installation positions of the two sensors illustrated in FIG. 2 by the delay circuit 361. The signal is delayed by the distance of the direction deviation M × pixel pitch and input to the second XY direction interpolation circuit 364. Then, the outputs of the first and second XY direction interpolation circuits 363 and 364 are input to the interpolation data merge circuit 365, and the interpolated sensor data 178 is output.

FIG. 20 is a schematic view illustrating the operation of the interpolation circuit used in the pattern defect inspection apparatus according to the fourth embodiment of the invention.
That is, FIGS. 20A and 20B illustrate the interpolation methods performed in the first and second XY interpolation circuits 363 and 364 illustrated in FIG. 19, respectively, and FIG. FIG. 20B illustrates an interpolation method of 2 × 2 pixels E to H in the second image data 181. FIG. 20B illustrates an interpolation method of 2 × 2 pixels A to D in the first image data 171. Yes.

  As illustrated in FIGS. 20A and 20B, the second image data 181 is delayed by the delay circuit 361 by the distance of M × pixel pitch, so that the 2 × of the first image data 171 is 2 ×. The positional relationship between 2 pixels A to D and 2 × 2 pixels E to H of the second image data 181 is shifted in the XY direction by 1 / N pixels (here, 1/2 pixels).

Then, the pixels ab, ac, bd, and cd at intermediate positions of the 2 × 2 pixels A to D of the first image data 171 are ab = (A + B) / 2, ac = (A + C) / 2, and bd, respectively. = (B + D) / 2 and cd = (C + D) / 2 are calculated as intermediate values of 2 × 2 pixels A to D, respectively.
In addition, the pixels ef, eg, fh, and gh in the middle of the 2 × 2 pixels E to H of the second image data 181 are ef = (E + F) / 2, eg = (E + G) / 2, and fh, respectively. = (F + H) / 2 and gh = (G + H) / 2 are calculated as intermediate values of 2 × 2 pixels E to H, respectively.

FIG. 21 is a schematic view illustrating a data merging method in the interpolation circuit used in the pattern defect inspection apparatus according to the fourth embodiment of the invention.
21 shows the pixel positions and interpolation data of the first image data 171 and the second image data 181 exemplified in FIGS. 20A and 20B, respectively. The figure on the right side of 21 is a diagram in which the respective interpolation data are superimposed. 21, the pixels A to D correspond to the pixel positions of the first image data 171, and the pixels E to H correspond to the pixel positions of the second image data 181.

  As illustrated in FIG. 21, the pixels α and β are positions corresponding to the pixels bd and cd of the first image data, and positions corresponding to the pixels ef and eg of the second image data. Accordingly, the merged value of the interpolation data of the pixels α and β is calculated as an average value of the interpolation data of the first and second image data 171 and 181. That is, α = (B + D + E + F) / 4 and β = (C + D + E + G) / 4.

FIG. 22 is a schematic diagram showing a specific example of an interpolation method and a data merge method in the interpolation circuit used in the pattern defect inspection apparatus according to the fourth embodiment of the present invention.
As shown in FIG. 22, the interpolation of the first and second image data 171 and 181 is performed by the method described in FIG. 20, and the merge of the interpolation data of the first and second image data 171 and 181 is as follows. The method described with reference to FIG. When pixel A = 100, pixel B = 60, pixel C = 60, pixel D = 40, pixel E = 60, pixel F = 20, pixel G = 20, and pixel H = 10, pixel α is (B + D + E + F) ) / 4 = (60 + 40 + 60 + 20) / 4 = 45, and the pixel β is (C + D + E + G) / 4 = (60 + 40 + 60 + 20) / 4 = 45.
Note that the blank pixel portion after interpolation merging illustrated in the right side of FIG. 22 includes the values of the pixels around the pixels A to D of the first image data 171 and the pixel E of the second image data 181. It can be calculated from the values of pixels around .about.H by the same method as described above.

  By interpolating the sensor data of the two sensors arranged at positions relatively shifted by 1 / N pixels (for example, 1/2 pixel) by the method described above, the effective pixel resolution can be maintained while maintaining the optical resolution. It is possible to obtain an image with a high pixel resolution that is N times (for example, twice) the pixel resolution. As a result, it is possible to avoid the problem of insufficient light quantity, sensor sensitivity, and operating speed caused by increasing the magnification of the optical system.

  As described above, in the pattern defect inspection apparatus 40 according to the fourth embodiment of the present invention, by providing the interpolation circuit 360, the data between the pixels can be accurately obtained as compared with the case where the normal averaging circuit 350 is used. By interpolating, further increasing the pixel resolution of the sensor, and eliminating noise due to radiation or the like, a pattern defect inspection apparatus with higher defect detection capability is provided.

(Fifth embodiment)
Next, a fifth embodiment of the present invention will be described. In the pattern defect inspection apparatus 10 according to the first embodiment, as illustrated in FIG. 5, the single event detection circuit 300 handles peripheral 2 × 2 pixel data, whereas in the pattern defect inspection apparatus according to the present embodiment. The single event detection circuit is different in that it handles data of 3 × 3 pixels in the vicinity. Since other parts can be the same as those in the first embodiment, only the single event detection circuit will be described below, and description of the other parts will be omitted.

FIG. 23 is a schematic view illustrating the configuration of a single event detection circuit used in a pattern defect inspection apparatus according to the fifth embodiment of the invention.
As shown in FIG. 23, the single event detection circuit 301 used in the pattern defect inspection apparatus according to the fifth embodiment of the present invention processes the first image data 171 and includes the first peripheral 3 × 3 pixels. An output unit 321, a first level difference calculation unit 322, a first maximum / minimum level difference calculation unit 324, and a first single event determination circuit 326 are included. Further, a second peripheral 3 × 3 pixel output unit 331, a second level difference calculation unit 332, a second maximum / minimum level difference calculation unit 334, and a second single event that process the second image data 181 A determination circuit 336 is included.

  The single event detection circuit 301 receives the first image data 171 and data obtained by delaying the second image data 181 by the delay circuit 310 by the distance of M × pixel pitch illustrated in FIG. . Then, it is detected whether there is a first single event that occurs in the first image data 171 and does not occur in the second image data 181, and a first single event detection flag 328 is output. Similarly, it is detected whether there is a second single event occurring in the second image data 181 and not occurring in the first image data 171, and a second single event detection flag 338 is output. To do.

FIG. 24 is a schematic view illustrating the operation of the single event detection circuit used in the pattern defect inspection apparatus according to the fifth embodiment of the invention.
FIG. 24 compares the target pixel 601 of the first image data 171 with the 3 × 3 pixel 605 around the target pixel 601 of the second image data 181, detects the first single event, This is a method of outputting as the single event detection flag 328. The figure also compares the target pixel 603 of the second image data 181 with the surrounding 3 × 3 pixel 606 of the target pixel 603 of the second image data 181 and detects the second single event. The method of outputting as the 2nd single event detection flag 338 is represented.

  As shown in FIG. 24, in order to detect a single event occurring in the first image data 171, first, the target pixel 601 of the first image data 171 and the position of the image of the target pixel 601 are determined. The first difference calculation unit 322 calculates the difference between the output levels of nine pixels of 3 × 3 pixels 605 around the second image data 181 as the center. Then, the first maximum / minimum level difference calculation unit 324 calculates the first maximum level difference MAX1 and the first minimum level difference MIN1.

  Next, the first single event determination circuit 326 determines whether the first maximum level difference MAX1 is larger than a preset first maximum threshold value. Further, it is determined whether the first minimum level difference MIN1 is smaller than a preset first minimum threshold value. And the sign of the difference between the first maximum level difference MAX1 and the first maximum threshold and the difference between the first minimum level difference MIN1 and the first minimum threshold are different, or The same is detected.

  Then, by the first logic result circuit 327, the logical sum of the threshold determination result of the first maximum level difference MAX1 and the threshold determination result of the first minimum level difference MIN1, and the result of the polarity inversion Compute the logical product of. As a result, the threshold value determination result of either the first maximum level difference MAX1 or the first minimum level difference MIN1 is valid, and the first maximum level difference MAX1 and the first maximum threshold value When the sign of the difference and the sign of the difference between the first minimum level difference MIN1 and the first minimum threshold value are the same, the first single event detection flag 328 is output.

  On the other hand, in order to detect a single event occurring in the second image data 181, first, the first pixel centered on the target pixel 603 of the second image data 181 and the image of the target pixel 603. The second difference calculation unit 332 calculates the difference between the output levels of nine pixels of the surrounding 3 × 3 pixels 606 of the image data 171. Then, the second maximum / minimum calculation unit 334 calculates the second maximum level difference MAX2 and the second minimum level difference MIN2.

  Next, the second single event determination circuit 336 determines whether the second maximum level difference MAX2 is larger than a preset second maximum threshold value. In addition, it is determined whether the second minimum level difference MIN2 is smaller than a preset second minimum threshold value. And the sign of the difference between the second maximum level difference MAX2 and the second maximum threshold and the difference between the second minimum level difference MIN2 and the second minimum threshold are different, or The same is detected.

  Then, by the second logic result circuit 337, the logical sum of the threshold value determination result of the second maximum level difference MAX2 and the threshold value determination result of the second minimum level difference MIN2, and the result of polarity inversion The logical product of is calculated. As a result, the result of the threshold value determination of either the second maximum level difference MAX2 or the second minimum level difference MIN2 is valid, and the second maximum level difference MAX2 and the second maximum threshold value When the sign of the difference and the sign of the difference between the second minimum level difference MIN2 and the second minimum threshold value are the same, the second single event detection flag 338 is output.

  As described above, in the pattern defect inspection apparatus according to the present embodiment, the single event detection circuit 301 can detect the single event even when the peripheral 3 × 3 pixel data is handled, and the pixel resolution of the sensor is increased. By eliminating the noise caused by the above, a pattern defect inspection apparatus having a high defect detection capability is provided.

  As described above, it is possible to effectively detect a single event based on data of peripheral Q × Q pixels (Q is an integer of 2 or more) around the target pixel.

  In the embodiment described above, the method of generating the interpolation data using the sensor in which the two sensors are relatively shifted by 0.5 pixels (1/2 pixel) has been described. Interpolation data may be generated using a plurality of sensors shifted by pixels (N is an integer of 1 or more). When an area sensor is used, interpolation data may be generated using two images obtained by shifting the scan position of the inspection object 150 by 1 / N of the pixel pitch.

(Sixth embodiment)
FIG. 25 is a schematic view illustrating the configuration of a pattern defect inspection apparatus according to the sixth embodiment of the invention.
As shown in FIG. 25, the pattern defect inspection apparatus 60 according to the sixth embodiment of the present invention is an example of a Die to Die inspection method, and the first image data 171 and the second image data 181 are used. Has an interpolation circuit 360 for interpolating. The sensor data 178 (interpolation data 362) is input to the reference data generation circuit 500 (data delay circuit).

  Thus, even in the pattern defect inspection apparatus 60 according to this embodiment of the Die to Die inspection method having the interpolation circuit 360, the defect detection capability is improved by increasing the pixel resolution of the sensor and eliminating noise due to radiation or the like. High pattern defect inspection apparatus can be provided.

(Seventh embodiment)
FIG. 26 is a flowchart illustrating the pattern defect inspection method according to the seventh embodiment of the invention.
As shown in FIG. 26, in the pattern defect inspection method according to the seventh embodiment of the present invention, first, at a plurality of first detection positions arranged at a first pitch along the first direction, The optical image 155 of the inspection object 150 is detected and the first image data 212 (171) is generated (step S110). In other words, the first image data 212 (171) is derived by capturing the optical image 155 of the inspection object 150.
For this, for example, the first sensor 210 composed of a TDI sensor or a line sensor illustrated in FIGS. 2A and 2B can be used. That is, the positions of the plurality of first pixels 211 of the first sensor 210 are the plurality of first detection positions. The plurality of first detection positions are arranged at a constant pixel pitch distance in the X direction, for example.

Next, the first detection positions are arranged in parallel in a second direction substantially orthogonal to the first direction as viewed from the plurality of first detection positions, arranged at a first pitch along the first direction, and first The plurality of second detection positions arranged at positions shifted by 1 / N (N is an integer of 1 or more) of the first pitch from the first detection position in the direction of The optical image 155 is detected and second image data 222 (181) is generated (step S120). That is, the second image data 222 (181) is derived by capturing the optical image 155 of the inspection object 150.
Similarly to this, the second sensor 220 including a TDI sensor or a line sensor can be used. That is, the positions of the plurality of second pixels 221 of the second sensor 220 are the plurality of second detection positions. As a result, the plurality of second detection positions are shifted from the first detection position by 1 / N pixel pitch.

Based on the first image data 212 (171) and the second image data 222 (181), a defect of the inspection object 150 is detected (step S130).
At this time, the above-described Die to Die inspection method (DD comparison) and Die to Database inspection method (DB comparison) can be used, and averaged data 352 of the first and second image data 171 and 181. Or interpolation data 362 can be used. In addition, it is possible to perform an inspection by excluding a single event that occurs only in one of the first and second image data 171 and 181.

  Thus, the pattern defect inspection method of the present embodiment provides a pattern defect inspection method with high defect detection capability by increasing the pixel resolution of the sensor and eliminating noise caused by radiation or the like.

  Note that the order of step S110 and step S120 illustrated in FIG. 26 is arbitrary, and step S110 and step S120 can be performed simultaneously.

  Further, the second detection position is from the first detection position to the second direction (direction (Y direction) substantially orthogonal to the arrangement direction (X-axis direction) of the first detection positions). The first pitch × (M + 1 / N) distance (M is an arbitrary integer) can be arranged at a shifted position.

(Eighth embodiment)
FIG. 27 is a flowchart illustrating a single event detection method performed in the pattern defect inspection method according to the eighth embodiment of the invention.
As already described, the single event is a first single event that occurs in the first image data 171 and does not occur in the second image data 181, and a single event that occurs in the second image data 181. The second single event that does not occur in the data 171 will be described below. The first single event will be described below.

  As shown in FIG. 27, in the single event detection method performed in the pattern defect inspection method according to the eighth embodiment of the present invention, first, one of the two image data is delayed (step S210). Here, the second image data 181 is delayed. At this time, the second image data 181 is delayed by the distance of the pixel pitch × M exemplified in FIG.

Then, a difference in output level between the target pixel 601 of the first image data 171 and the peripheral Q × Q pixel 602 of the second image data 181 centering on the position of the image of the target pixel 601 is calculated (step) S220).
Here, Q is an integer equal to or greater than 2. For example, data processing when Q is 2 is performed as described in FIGS. 5 and 6, and data processing when Q is 3 is as shown in FIG. The operation is performed as described in FIG.

  Then, the first maximum level difference MAX1 and the first minimum level difference MIN1 are calculated (step S230).

Then, the magnitude of the first maximum level difference MAX1 with respect to the preset first maximum threshold is determined, and the magnitude of the first minimum level difference MIN1 with respect to the preset first minimum threshold is determined. Is determined (step S240).
And the sign of the difference between the first maximum level difference MAX1 and the first maximum threshold and the difference between the first minimum level difference MIN1 and the first minimum threshold are different, or Detect if they are the same.

Then, the logical product of the logical sum of the threshold determination result of the first maximum level difference MAX1 and the threshold determination result of the first minimum level difference MIN1 and the result of polarity inversion is calculated (step S250). ).
As a result, the threshold value determination result of either the first maximum level difference MAX1 or the first minimum level difference MIN1 is valid, and the first maximum level difference MAX1 and the first maximum threshold value A first single event is detected when the sign of the difference and the sign of the difference between the first minimum level difference MIN1 and the first minimum threshold are the same.
The second single event can also be detected by the same method as described above.

  Thereby, by the pattern defect inspection method of this embodiment, noise due to radiation or the like can be efficiently eliminated, and a pattern defect inspection method having high defect detection capability is provided.

  Note that the order of steps S210 to S250 illustrated in FIG. 27 is arbitrary as long as technically possible, and any two or more of steps S210 to S250 can be performed simultaneously.

(Ninth embodiment)
FIG. 28 is a flowchart illustrating an image data interpolation method performed in the pattern defect inspection method according to the ninth embodiment of the invention.
As shown in FIG. 28, in the image data interpolation method performed by the pattern defect inspection method according to the ninth embodiment of the present invention, first, interpolation data is obtained from the first image data 171 (step S310). For example, the method illustrated in FIG. 20 can be used.

  Then, interpolation data is obtained from the second image data 181 (step S320). Also for this, for example, the method illustrated in FIG. 20 can be used.

  Then, the interpolation data obtained from the first image data 171 and the interpolation data obtained from the second image data 181 are merged (step S330). For example, the method illustrated in FIG. 21 can be used.

  Thereby, the pattern defect inspection method according to the present embodiment can accurately interpolate data between pixels (between detection positions), increase the pixel resolution of the sensor, and provide a pattern defect inspection apparatus with high defect detection capability. Provided.

  Note that the pattern defect inspection apparatus and inspection method according to the embodiments of the present invention described above are not limited to photomasks used in semiconductor devices, but various fine structures such as MEMS (Micro-electro-mechanical System) and magnetic elements. It can be used for inspection of pattern defects of various photomasks used for manufacturing a body.

The embodiments of the present invention have been described above with reference to specific examples. However, the present invention is not limited to these specific examples. For example, regarding the specific configuration of each element constituting the pattern defect inspection apparatus and the inspection method, those skilled in the art can implement the present invention in the same manner by appropriately selecting from a known range, and the same effect can be obtained. As long as it is within the scope of the present invention.
Moreover, what combined any two or more elements of each specific example in the technically possible range is also included in the scope of the present invention as long as the gist of the present invention is included.
In addition, based on the pattern defect inspection apparatus and inspection method described above as an embodiment of the present invention, all pattern defect inspection apparatuses and inspection methods that can be implemented by those skilled in the art as appropriate are included in the gist of the present invention. As long as it is included, it belongs to the scope of the present invention.
In addition, in the category of the idea of the present invention, those skilled in the art can conceive various changes and modifications, and it is understood that these changes and modifications also belong to the scope of the present invention. .

It is a schematic diagram which illustrates the structure of the pattern defect inspection apparatus which concerns on the 1st Embodiment of this invention. 1 is a schematic plan view illustrating the configuration of a sensor used in a pattern defect inspection apparatus according to a first embodiment of the invention. 1 is a schematic plan view illustrating the configuration of a sensor used in a pattern defect inspection apparatus according to a first embodiment of the invention. 1 is a schematic plan view illustrating the configuration of a sensor used in a pattern defect inspection apparatus according to a first embodiment of the invention. It is a schematic diagram which illustrates the structure of the single event detection circuit used for the pattern defect inspection apparatus which concerns on the 1st Embodiment of this invention. It is a schematic diagram which illustrates operation | movement of the single event detection circuit used for the pattern defect inspection apparatus which concerns on the 1st Embodiment of this invention. It is a schematic diagram which illustrates operation | movement of the single event detection circuit used for the pattern defect inspection apparatus which concerns on the 1st Embodiment of this invention. It is a schematic diagram which illustrates operation | movement of the single event detection circuit used for the pattern defect inspection apparatus which concerns on the 1st Embodiment of this invention. It is a schematic diagram which illustrates operation | movement of the single event detection circuit used for the pattern defect inspection apparatus which concerns on the 1st Embodiment of this invention. It is a schematic diagram which illustrates the structure of the defect detection part used for the pattern defect inspection apparatus which concerns on the 1st Embodiment of this invention. It is a schematic diagram which illustrates the structure of the differential comparison circuit used for the pattern defect inspection apparatus which concerns on the 1st Embodiment of this invention. It is a schematic diagram which illustrates the structure of the pattern defect inspection apparatus of a comparative example. It is a schematic diagram which illustrates the structure of the defect detection part used for the pattern defect inspection apparatus of a comparative example. It is a model top view which illustrates the structure of the sensor used for the pattern defect inspection apparatus which concerns on the 2nd Embodiment of this invention. It is a model top view which illustrates operation | movement of the sensor used for the pattern defect inspection apparatus which concerns on the 2nd Embodiment of this invention. It is a schematic diagram which illustrates the structure of the pattern defect inspection apparatus which concerns on the 3rd Embodiment of this invention. It is a schematic diagram which illustrates the inspection method of the pattern defect inspection apparatus which concerns on embodiment of this invention. It is a schematic diagram which illustrates the structure of the pattern defect inspection apparatus which concerns on the 4th Embodiment of this invention. It is a schematic diagram which illustrates the structure of the interpolation circuit used for the pattern defect inspection apparatus which concerns on the 4th Embodiment of this invention. It is a schematic diagram which illustrates operation | movement of the interpolation circuit used for the pattern defect inspection apparatus which concerns on the 4th Embodiment of this invention. It is a schematic diagram which illustrates the data merge method in the interpolation circuit used for the pattern defect inspection apparatus which concerns on the 4th Embodiment of this invention. It is a schematic diagram which shows the specific example of the interpolation method and data merge method in the interpolation circuit used for the pattern defect inspection apparatus which concerns on the 4th Embodiment of this invention. It is a schematic diagram which illustrates the structure of the single event detection circuit used for the pattern defect inspection apparatus which concerns on the 5th Embodiment of this invention. It is a schematic diagram which illustrates operation | movement of the single event detection circuit used for the pattern defect inspection apparatus which concerns on the 5th Embodiment of this invention. It is a schematic diagram which illustrates the structure of the pattern defect inspection apparatus which concerns on the 6th Embodiment of this invention. It is a flowchart figure which illustrates the pattern defect inspection method which concerns on the 7th Embodiment of this invention. It is a flowchart figure which illustrates the detection method of the single event performed with the pattern defect inspection method which concerns on the 8th Embodiment of this invention. It is a flowchart figure which illustrates the interpolation method of the image data performed with the pattern defect inspection method which concerns on the 9th Embodiment of this invention.

Explanation of symbols

10, 30, 40, 60, 91 Pattern defect inspection apparatus 110 Laser light source 120 Illumination optical system 122 Illumination light 130 Illumination unit 140 Imaging optical system 150 Inspected object (photomask)
155 Optical image 160 XY table 170 First AD conversion circuit 171, 212 First image data 178, 198 Sensor data 180 Second AD conversion circuit 181, 222 Second image data 200, 290 Sensor 201 Up direction (back) direction)
202 downward (front)
210 First sensor 211 First pixel 213 First image digital data 220 Second sensor 221 Second pixel 223 Second image digital data 230 TDI sensor 231 Light receiving unit (pixel)
232 Vertical transfer register 233, 234 Horizontal transfer register 250 First line sensor (first sensor)
251 First line sensor pixel (first pixel)
252 Second line sensor (second sensor)
253 Second line sensor pixel (second pixel)
261 to 268, 271 to 274 Output unit 281 and 282 Multiplexer (MUX)
291, 292 Output 300, 301 Single event detection circuit 302 Single event data (single event detection flag)
310 delay circuit 320 first peripheral 2 × 2 pixel output unit 321 first peripheral 3 × 3 pixel output unit 322 first level difference calculation unit 324 first maximum / minimum level difference calculation unit 326 first single event Determination circuit 327 First logic result circuit 328 First single event transfer flag 330 Second peripheral 2 × 2 pixel output unit 331 Second peripheral 3 × 3 pixel output unit 332 Second level difference calculation unit 334 Second Maximum / minimum level difference calculation unit 336 second single event determination circuit 337 second logic result circuit 338 second single event transfer flag 350 averaging circuit 352 averaged data 360 interpolation circuit 361 delay circuit 363 first XY Direction interpolation circuit 364 Second XY direction interpolation circuit 365 Interpolation data merge circuit 400, 490 Defect detection unit 410 Alignment circuit 420 Level comparison circuit 430 Differential comparison circuit 431 Differentiation circuit 432 Selector 433 Edge direction detection circuit 434 Peripheral pixel edge direction differentiation circuit 435 Maximum value detection circuit 436 Comparison operation circuit 438 Differential comparison data 440, 450 Defect determination circuit 460 OR circuit 470 AND circuit 480 Pattern Defect inspection result 500, 501 Reference data generation circuit 510 Design data 520 Data expansion circuit 521 Data expansion data 522 Reference data 530 Laser interferometer 532 Position data 601, 603 Target pixel 602, 604 Peripheral 2 × 2 pixels 605, 606 Peripheral 3 × 3 pixels 620 Noise event part 630 Defective part 640 Edge part

Claims (11)

A plurality of first pixels arranged at a first pixel pitch along a first direction, and an optical image of the object to be inspected is converted into an electrical image signal to output first image data; 1 sensor,
The first sensor is arranged in parallel in a second direction substantially orthogonal to the first direction as viewed from the first sensor, arranged at the first pixel pitch along the first direction, and the first direction. And a plurality of second pixels arranged at positions shifted by 1 / N (N is an integer of 1 or more) of the first pixel pitch from the arrangement position of the first pixels, A second sensor that converts the optical image of the second image into an electrical image signal and outputs second image data;
A defect detection unit that detects a defect of the inspection object based on the first image data and the second image data;
A pattern defect inspection apparatus comprising:   The second pixel is arranged at a position shifted from the arrangement position of the first pixel by the first pixel pitch × (M + 1 / N) (M is an arbitrary integer) in the second direction. The pattern defect inspection apparatus according to claim 1, wherein:   A first single event that occurs in the first image data and does not occur in the second image data is detected, and a second event that occurs in the second image data and does not occur in the first image data. The pattern defect inspection apparatus according to claim 1, further comprising a single event detection circuit that detects a single event.   The defect detection unit includes a detection result of the first single event, a detection result of the second single event, a first defect detection result of the first image data, and the second image data. 4. The pattern defect inspection apparatus according to claim 3, further comprising a logic result circuit for deriving a logic result of the second defect detection result.   The first sensor and the second sensor are TDI sensors, and the first image data and the second image data can be output from a common output terminal. The pattern defect inspection apparatus as described in any one of 1-4.   6. The multiplexer according to claim 1, further comprising a multiplexer connected to the first readout unit of the first sensor and the second readout unit of the second sensor. The pattern defect inspection apparatus described in 1.   A data interpolation circuit for obtaining interpolation data between the first pixels and between the second pixels using the first image data and the second image data; The pattern defect inspection apparatus according to claim 1, wherein the pattern defect inspection apparatus is a pattern defect inspection apparatus. Detecting an optical image of the object to be inspected at a plurality of first detection positions arranged at a first pitch along a first direction to generate first image data;
The first detection positions are arranged in parallel in a second direction substantially orthogonal to the first direction, arranged at the first pitch along the first direction, and the first The optical image of the object to be inspected at a plurality of second detection positions arranged at positions shifted by 1 / N (N is an integer of 1 or more) of the first pitch from the first detection position in the direction of. To generate second image data,
A pattern defect inspection method, comprising: detecting a defect of the inspection object based on the first image data and the second image data.   The second detection position is arranged at a position shifted from the first detection position by the first pitch × (M + 1 / N) (M is an arbitrary integer) in the second direction. The pattern defect inspection method according to claim 8.   The detection of the defect occurs in the first image data and does not occur in the second image data, and in the first image data and occurs in the second image data. The pattern defect inspection method according to claim 8, further comprising excluding a second single event that is not performed.   11. The detection of the defect includes performing the detection using interpolation data interpolated using the first image data and the second image data. The pattern defect inspection method as described in 2.

Images