JP2023073914A - Mask inspection device and mask inspection method - Google Patents

Abstract

PURPOSE: To provide an inspection device capable of measuring a change in an uneven distribution of illumination light.CONSTITUTION: A mask inspection device 100 includes a beam splitter 190 that splits inspection light, a TDI sensor (imaging sensor 105) that captures an image of a mask substrate obtained by irradiating the mask substrate on which a pattern is formed with first light split from the inspection light and outputs image data, a line sensor 194 that receives second light split from the inspection light and measures a change over time in a light intensity distribution in a direction orthogonal to the accumulation direction of the TDI sensor, a correction circuit 117 that corrects the image data output from the TDI sensor in accordance with the location of the light intensity distribution using the value of the light intensity distribution that changes over time, a comparison circuit 108 that compares an image to be inspected using the corrected image data with a predetermined image.SELECTED DRAWING: Figure 1

Description

The present invention relates to a mask inspection apparatus and a mask inspection method. For example, it relates to an inspection apparatus for inspecting a pattern defect of a mask substrate.

2. Description of the Related Art In recent years, as large-scale integrated circuits (LSIs) have become highly integrated and have large capacities, the circuit line width required for semiconductor elements has become increasingly narrow. These semiconductor elements are manufactured by exposing and transferring the pattern onto a wafer using a reduction projection exposure apparatus called a stepper using an original image pattern (also called a mask or reticle, hereinafter collectively referred to as a mask) on which a circuit pattern is formed. Manufactured by circuit forming.

In addition, the improvement of yield is essential for the manufacture of LSIs, which requires a great manufacturing cost. One of the major factors that reduce the yield is pattern defects such as shape defects and/or dimensional defects of the mask used when exposing and transferring ultrafine patterns onto semiconductor wafers by photolithography technology. be done. In recent years, as the dimensions of LSI patterns formed on semiconductor wafers have become finer, the dimensions that must be detected as pattern defects have become extremely small. Therefore, it is necessary to improve the precision of pattern inspection apparatuses for inspecting defects in transfer masks used in LSI manufacturing.

As an inspection method, for example, a "die to die inspection" that compares optical image data obtained by imaging the same pattern at different locations on the same mask, and a pattern designing CAD data as a mask. The drawing data (design data) converted into the device input format for input by the drawing device at the time of drawing is input to the inspection device, and based on this, a reference image is generated, and it becomes measurement data that captures the pattern. There is a "die to database inspection" that compares optical images.

Here, the light amount of the laser light, which is the inspection light for imaging the pattern on the mask, changes with time. Conventionally, changes in the amount of light over time have been measured by splitting a portion of the inspection light in an illumination system, condensing the light with a lens, collecting the entire luminous flux, and measuring the light amount of the entire luminous flux with a sensor. A method of correcting pixel values of an image using changes in the measured light intensity of the entire luminous flux has been studied (see, for example, Patent Document 1). However, the amount of light is distributed rather than uniquely in the entire illumination field. Such a change in distribution unevenness of the illumination causes an error in the captured image. With this measurement method, it was difficult to monitor changes in the uneven distribution of the illumination field at the mask conjugate position. Therefore, there is a problem that even if the light amount of the total luminous flux is measured, the change according to the position of the light amount distribution cannot be detected. As a method for correcting the amount of light, a method of measuring the amount of light in a part of the image sensor is being considered, but it is difficult to follow changes in the uneven distribution of the illumination light, and another method is required. .

JP 2006-250944 A

Accordingly, embodiments of the present invention provide an inspection apparatus and method capable of measuring changes in distribution unevenness of illumination light.

A mask inspection apparatus according to one aspect of the present invention includes:
a beam splitter that splits the inspection light;
a time delay integration (TDI) sensor that captures an image of a mask substrate obtained by irradiating a mask substrate having a pattern formed thereon with a first light branched from inspection light and outputs image data;
a line sensor that receives the second branched light of the inspection light and measures a temporal change in the light amount distribution in the direction orthogonal to the accumulation direction of the TDI sensor;
a correction unit that corrects the image data output from the TDI sensor according to the position of the light intensity distribution using the value of the light intensity distribution that changes over time;
a comparison unit that compares the image to be inspected using the corrected image data with a predetermined image;
characterized by comprising

Further, it is preferable that the correction unit corrects the image data at the target position using the value of the position of the light amount distribution corresponding to the target position of the image data in terms of the mask surface.

Further, it is preferable that the correcting unit corrects the image data of the target pixel using the value of the light quantity distribution measured during the time period when the image of the target pixel of the mask substrate is captured by the TDI sensor.

Also, the inspection light is preferably illuminated using Koehler illumination.

Further, it is preferable that the correcting unit corrects the image data of the target pixel using a moving average value of each value of the light intensity distribution measured during the time period when the image of the target pixel on the mask substrate is captured by the TDI sensor. .

A mask inspection method according to one aspect of the present invention includes:
a step of branching the inspection light;
A step of capturing an image of the mask substrate obtained by irradiating the mask substrate on which the pattern is formed with the first branched light of the inspection light using a time delay integration (TDI) sensor, and outputting image data. and,
a step of using a line sensor to receive the second light branched from the inspection light and measuring a temporal change in the light amount distribution in a direction orthogonal to the accumulation direction of the TDI sensor;
correcting the image data output from the TDI sensor according to the position of the light intensity distribution using the value of the light intensity distribution that changes with time;
comparing the image under inspection using the corrected image data to a predetermined image and outputting the results;
characterized by comprising

According to the embodiments of the present invention, changes in distribution unevenness of illumination light can be measured. Therefore, the change according to the position of the distribution unevenness can be corrected. As a result, the reproducibility of the device can be improved.

1 is a configuration diagram showing the configuration of a pattern inspection apparatus according to Embodiment 1; FIG. FIG. 2 is a conceptual diagram for explaining an inspection area according to Embodiment 1; FIG. FIG. 4 is a diagram for explaining a temporal change in light amount distribution in Embodiment 1; FIG. 4 is a diagram for explaining a method of measuring distribution unevenness of illumination according to Embodiment 1; FIG. FIG. 10 is a diagram for explaining light amount measurement in a comparative example of the first embodiment; 4A and 4B are diagrams for explaining light intensity measurement in Embodiment 1. FIG. FIG. 2 is a flow chart showing main steps of an inspection method according to Embodiment 1; FIG. 4 is a diagram for explaining a correction method according to the first embodiment; FIG. 4 is a diagram for explaining the relationship between magnification and image size at each position in Embodiment 1. FIG. 4 is a diagram for explaining the relationship between each position of the light amount distribution and the position of the image of the TDI sensor in Embodiment 1. FIG. FIG. 4 is a diagram for explaining filter processing according to Embodiment 1; FIG. 2 is a diagram showing an example of an internal configuration of a comparison circuit according to Embodiment 1; FIG.

Embodiment 1.
FIG. 1 is a configuration diagram showing the configuration of the pattern inspection apparatus according to the first embodiment. In FIG. 1, an inspection apparatus 100 for inspecting pattern defects formed on a substrate 101 (an example of a substrate to be inspected) includes an optical image acquisition mechanism 150 and a control system circuit 160 (control unit).

The optical image acquisition mechanism 150 includes a light source 103 that generates laser light, a Koehler illumination optical system 300, a relay lens 31, a beam splitter 32, a transmission illumination optical system 170, a reflection illumination optical system 270, an XYθ table 102, an objective lens 104, a beam Splitter 176, mirror 177, imaging optical system 178, imaging optical system 278, imaging sensor 105, sensor circuit 106, correction circuit 117, stripe pattern memory 123, imaging sensor 205, sensor circuit 206, correction circuit 217, stripe pattern memory 223 , a beam splitter 190 , a cylinder lens 192 , a line sensor 194 , a memory 196 , a beam splitter 290 , a cylinder lens 292 , a line sensor 294 and a memory 296 .

The transmitted illumination optical system 170 is composed of one or more lenses and/or one or more mirrors. In the example of FIG. 1, transillumination optics 170 includes mirror 134 , mirror 138 and lens 171 .

The reflected illumination optical system 270 is composed of one or more lenses and/or one or more mirrors. In the example of FIG. 1, the reflected illumination optics 270 has a lens 271 .

A substrate 101 transferred from an autoloader (not shown) is placed on the XYθ table 102 . The substrate 101 includes, for example, an exposure photomask for transferring a pattern to a semiconductor substrate such as a wafer. A plurality of graphic patterns to be inspected are formed on this photomask. The substrate 101 is placed, for example, on the XYθ table 102 with the pattern formation surface facing downward.

As the imaging sensors 105 and 205, for example, TDI (time delay integration) sensors are preferably used. A TDI sensor has a plurality of photosensor elements arranged two-dimensionally. Each photosensor element is set with a predetermined image accumulation time (or sometimes referred to as a scan time; the same shall apply hereinafter) when capturing an image. The TDI sensor integrates and outputs the outputs of a plurality of photosensor elements arranged in the scanning direction. A plurality of photosensor elements arranged in the scanning direction pick up images of the same pixels while shifting the time according to the movement of the XYθ table 102 .

Beam splitter 190 is positioned within transillumination optics 170 . The line sensor 194 is arranged at a position conjugate with the pattern formation surface of the substrate 101 . Beam splitter 290 is positioned within reflective illumination optics 270 . The line sensor 294 is arranged at a position conjugate with the pattern formation surface of the substrate 101 .

The line sensors 194 and 294 have a plurality of photosensor elements arranged in a direction perpendicular to the scanning direction (eg Y direction). The number of photosensor elements of the line sensors 194 and 294 may be the same as or different from the number of photosensor elements arranged in a direction perpendicular to the scanning direction of the imaging sensors 105 and 205 (for example, the Y direction). I don't mind.

Koehler illumination optics 300 includes, for example, beam expander 302 , rotating phase plate 306 , splitting lens 304 , collimating lens 308 , and illumination slit 310 .

In the control system circuit 160, the control computer 110 that controls the inspection apparatus 100 as a whole is connected via the bus 120 to the magnetic disk device 109, the memory 111, the position circuit 107, the comparison circuit 108, the reference image creation circuit 112, and the table control circuit. 114.

The correction circuit 117 is connected to the stripe pattern memory 123 and the correction circuit 217 is connected to the stripe pattern memory 223 . The stripe pattern memories 123 and 223 are connected to the comparator circuit 108 . The function of the correction circuit 117 is preferably installed in the sensor circuit 106 as well. Similarly, correction circuit 217 may preferably be implemented with its functionality within sensor circuit 206 .

The output of line sensor 194 is connected to memory 196 and the output of line sensor 294 is connected to memory 296 . Memory 196 is connected to correction circuit 117 . Memory 296 is connected to correction circuit 217 .

Also, the XYθ table 102 is driven by a drive mechanism 115 . The drive mechanism 115 has, for example, an X-axis motor, a Y-axis motor, and a θ-axis motor, and the XYθ table 102 is driven by the X-axis motor, Y-axis motor, and θ-axis motor. The XYθ table 102 is an example of a stage. The reference image generation circuit 112 is also connected to the comparison circuit 108 .

Linear motors, for example, can be used for these X-axis motor, Y-axis motor, and θ-axis motor. The XYθ table 102 can be moved in horizontal and rotational directions by motors for the XYθ axes. The XYθ table 102 is adjusted to the focal position (optical axis direction: Z-axis direction) between the pattern forming surface of the substrate 101 and the imaging sensors 105 and 205 under the control of the control computer 110 . The movement position of the substrate 101 placed on the XYθ table 102 is measured by a laser length measurement system (not shown) and supplied to the position circuit 107 .

A series of "circuits" such as the position circuit 107, the comparison circuit 108, the reference image generation circuit 112, and the table control circuit 114 have processing circuits. Such processing circuits include electrical circuits, computers, processors, circuit boards, quantum circuits, or semiconductor devices. Alternatively, different processing circuits (separate processing circuits) may be used. For example, a series of “circuits” such as comparison circuit 108 , reference image generation circuit 112 , and table control circuit 114 may be configured and executed by control computer 110 . Input data required for the comparison circuit 108, the reference image generation circuit 112, and the table control circuit 114 or the results of calculations are stored in a memory (not shown) or the memory 111 in each circuit each time. Input data necessary for the control computer 110 or results of calculation are stored in a memory (not shown) in the control computer 110 or the memory 111 each time. A program that causes a computer, processor, or the like to be executed may be recorded on a recording medium such as the magnetic disk device 109 or the like.

In inspection apparatus 100, light source 103, Koehler illumination optical system 300, relay lens 31, beam splitter 32, transmission illumination optical system 170, objective lens 104, beam splitter 176, mirror 177, imaging optical system 178, imaging sensor 105, sensor The circuit 106 and the correction circuit 117 constitute a transmission inspection optical system with a high magnification M1. As the magnification M1, for example, an inspection optical system with a magnification of 200 to 300 is configured.

Similarly, in the inspection apparatus 100, the light source 103, the Koehler illumination optical system 300, the relay lens 31, the beam splitter 32, the reflected illumination optical system 270, the beam splitter 176, the objective lens 104, the imaging optical system 278, the imaging sensor 205, the sensor The circuit 206 and the correction circuit 217 constitute a reflection inspection optical system with a high magnification M1. As the magnification M1, for example, an inspection optical system with a magnification of 200 to 300 is configured.

Drawing data (design data), which is the basis for pattern formation on the substrate to be inspected 101 , is input from the outside of the inspection apparatus 100 and stored in the magnetic disk device 109 . A plurality of figure patterns are defined in the drawing data, and each figure pattern is usually composed of a combination of a plurality of element figures. Note that there may be a figure pattern composed of one figure. A corresponding pattern is formed on the substrate 101 to be inspected based on each figure pattern defined by the drawing data.

Here, in FIG. 1, constituent parts necessary for explaining the first embodiment are described. It goes without saying that other configurations normally required for the inspection apparatus 100 may be included.

The light source 103 emits, for example, laser light (for example, DUV light) having a wavelength of about 190 to 200 nm (an example of ultraviolet light) as inspection illumination light. By making the laser beam 301 emitted from the light source 103 incident on the Koehler illumination optical system 300, ideally, inspection light capable of uniformly illuminating the surface is generated. Specifically, it operates as follows. A laser beam 301 emitted from a light source 103 is expanded by a beam expander 302 and passed through a rotating phase plate 306 to reduce spatial coherency. Light passing through the rotating phase plate 306 generates a surface light source at the split lens 304 . After that, a collimator lens 308 generates a Fourier plane of the surface light source, and an illumination slit 310 is arranged at the position of this Fourier plane.

The inspection light that has passed through the illumination slit 310 is projected onto the beam splitter 32 by the relay lens 31 . The beam splitter 32 splits the inspection light into light for transmitted illumination and light for reflected illumination. When reflection inspection is not performed, the light branched to the reflection illumination optical system 270 may be shielded. Conversely, when transmission inspection is not performed, the light branched to the transmission illumination optical system 170 may be blocked. In FIG. 1, the inspection light for transmission inspection and the image separated by the beam splitter 32 are indicated by dotted lines. In FIG. 1, the inspection light and image for reflection inspection separated by the beam splitter 32 are indicated by solid lines.

In transmission inspection, inspection light for transmission inspection separated by the beam splitter 32 is reflected by the mirror 134 and enters the beam splitter 190 . The inspection light for transmission inspection is split into two optical paths by the beam splitter 190 . For example, 0.1 to 10% (for example, 1%) of the inspection light for transmission inspection is separated toward the cylinder lens 192 as correction light. The remainder of the inspection light becomes transmitted illumination light. Of the inspection light, transmitted illumination light (first light) split by the beam splitter 190 is reflected by the mirror 138 and illuminates the substrate 101 through the lens 171 . Transmitted light that has passed through the substrate 101 passes through the beam splitter 176 via the objective lens 104 . Then, it is reflected by the mirror 177, formed as an optical image (transmission image) by the imaging optical system 178 on the imaging sensor 105, and enters. In this manner, the imaging sensor 105 captures a transmission image.

A pattern image formed on the image sensor 105 is photoelectrically converted by each photosensor element of the image sensor 105 , and a value after integration of a plurality of photosensor elements arranged in the scanning direction is output to the sensor circuit 106 . Then, it is A/D (analog/digital) converted by the sensor circuit 106 .

Alternatively/and in reflection inspection, the inspection light for reflection inspection separated by the beam splitter 32 enters the beam splitter 290 . The inspection light for reflection inspection is split into two optical paths by the beam splitter 290 . For example, 0.1 to 10% (for example, 1%) of the inspection light for reflection inspection is separated toward the cylinder lens 292 as correction light. The remainder of the inspection light becomes the reflected illumination light. Of the inspection light, the reflected illumination light (second light) split by the beam splitter 290 is reflected by the beam splitter 176 and irradiated onto the substrate 101 by the objective lens 104 . In other words, the illumination optical system composed of the reflective illumination optical system 170, the beam splitter 176, and the objective lens 104 illuminates the patterned substrate 101 to be inspected. Reflected light from the substrate 101 passes through the objective lens 104 and the beam splitter 174, is formed as an optical image (reflected image) by the imaging optical system 278 on the imaging sensor 205, and enters the imaging sensor 205. FIG. In this manner, the imaging sensor 205 captures a reflected image.

A pattern image formed on the image sensor 205 is photoelectrically converted by each photosensor element of the image sensor 205 , and a value after integration of a plurality of photosensor elements arranged in the scanning direction is output to the sensor circuit 206 . Then, it is A/D (analog/digital) converted by the sensor circuit 206 .

FIG. 2 is a conceptual diagram for explaining inspection areas in the first embodiment. As shown in FIG. 2, the inspection area 10 (entire inspection area) of the substrate 101 is virtually divided into a plurality of strip-shaped inspection stripes 20 having a scanning width W of the TDI sensor 105 in the Y direction, for example. . Then, the inspection apparatus 100 acquires an image (stripe area image) for each inspection stripe 20 . For each inspection stripe 20, a laser beam (inspection light) is used to pick up an image of a graphic pattern arranged in the inspection stripe 20 in the longitudinal direction (X direction) of the stripe region. In order to prevent an image from being lost, it is preferable that the plurality of inspection stripes 20 be set so that adjacent inspection stripes 20 overlap each other with a predetermined margin width.

As the XYθ table 102 moves, the TDI sensor 105 relatively continuously moves in the X direction to acquire an optical image. The TDI sensor 105 continuously captures optical images with a scan width W as shown in FIG. In other words, the TDI sensor 105 picks up an optical image on the surface of the substrate 101 on which a plurality of figure patterns are formed while relatively moving in the integration direction of the TDI sensor 105 . In the first embodiment, after an optical image of one inspection stripe 20 is captured, the optical image of the scan width W is similarly captured while moving in the Y direction to the position of the next inspection stripe 20 and then moving in the opposite direction. Take images continuously. That is, imaging is repeated in the forward (FWD)-back forward (BWD) directions in opposite directions on the outbound and return trips.

In the actual inspection, the stripe area image of each inspection stripe 20 is divided into images of a plurality of rectangular frame areas 30, as shown in FIG. Then, each image in the frame area 30 is inspected. For example, it is divided into a size of 1024×1024 pixels. Therefore, a reference image to be compared with the frame image 31 of the frame area 30 is similarly created for each frame area 30 .

Here, the imaging direction is not limited to repetition of forward (FWD)-back forward (BWD). You may image from one direction. For example, FWD-FWD may be repeated. Alternatively, BWD-BWD may be repeated.

3A and 3B are diagrams for explaining temporal changes in the light amount distribution according to the first embodiment. FIG. The example of FIG. 3 shows the case where the imaging sensor 105 (205) images the inspection stripe 20 located at Y=Y0 to Y1. As shown in FIG. 3, there is a light amount distribution of the inspection light in the direction (Y direction) orthogonal to the scanning direction in the area A arranged in the scanning direction. As scanning progresses, when scanning area B, the light intensity distribution of the inspection light in the direction (Y direction) perpendicular to the scanning direction of area B is different from the inspection light amount distribution in the direction (Y direction) perpendicular to the scanning direction of area A. It differs from the light quantity distribution of light. In other words, the light amount distribution of the inspection light changes with time. In this way, the amount of light is not unique but distributed over the entire illumination field. Such a change in distribution unevenness of the illumination causes an error in the captured image. With the conventional measurement method of measuring the amount of light by the total luminous flux, it was difficult to monitor changes in the uneven distribution of the illumination field at the mask conjugate position. Therefore, there is a problem that even if the light amount of the total luminous flux is measured, the change according to the position of the light amount distribution cannot be detected. As a method for correcting the amount of light, a method of measuring the amount of light in a part of the image sensor is being considered, but it is difficult to follow changes in the uneven distribution of the illumination light, and another method is required. . Therefore, in Embodiment 1, as described with reference to FIG. 1, the line sensor 194 (294) arranged at the mask conjugate position measures the change in the illumination distribution unevenness.

4A and 4B are diagrams for explaining a method of measuring the distribution unevenness of illumination according to the first embodiment. FIG. In the example of FIG. 4, illustration of the illumination optical system after the Koehler illumination optical system 300 is omitted. Although the Koehler illumination optical system 300 ideally controls the inspection light to have an in-plane uniform light intensity, a light intensity distribution occurs on the substrate 101 surface as shown in FIG. Furthermore, the light intensity distribution of the laser light emitted from the light source 103 changes with time due to, for example, changes in beam diameter and optical axis. Therefore, in Embodiment 1, the line sensor 194 measures the light quantity distribution in the direction orthogonal to the scanning direction of the imaging sensor 105 .

FIG. 5 is a diagram for explaining light amount measurement in a comparative example of the first embodiment. The comparative example of FIG. 5 shows a case where a spherical lens is used as a condensing lens for condensing the illumination light 11 . Therefore, the entire luminous flux of the illumination light 11 is condensed at one point, and the condensed light is measured by one photosensor 300, for example. In such a case, it becomes difficult to acquire the light amount distribution from the obtained data.

6A and 6B are diagrams for explaining light intensity measurement in the first embodiment. FIG. As shown in FIG. 6, in the first embodiment, the illumination light 11 (second light: light for correction) split by the beam splitter 190 (290) out of the inspection light is condensed into a condensing lens having a circular shape. A columnar or semi-cylindrical cylinder lens 192 (292) is used. The cylinder lens 192 (292) is arranged so that the generatrix direction (extending direction) of the cylinder lens 192 (292) is aligned with the direction perpendicular to the scanning direction of the imaging sensor 105 (205). Similarly, the longitudinal direction of the line sensor 194 (294) is arranged so as to correspond to the direction orthogonal to the scanning direction. As a result, when the divided illumination light 11 is condensed by the cylinder lens 192 (292), the light spread in the direction in which the cylinder lens 192 (292) extends in the plane perpendicular to the traveling direction of the illumination light 11 is Light that spreads in a direction perpendicular to the extending direction of the cylinder lens 192 (292) without condensing is condensed to one point. Light passing through the cylinder lens 192 (292) is measured by a plurality of photosensor elements arranged in the longitudinal direction of the line sensor 194 (294). Therefore, it is possible to measure the light amount distribution in the longitudinal direction of the line sensor 194 (294), which is focused only in the lateral direction of the line sensor 194 (294). The longitudinal direction of the line sensor 194 (294) is a direction perpendicular to the scanning direction of the imaging sensor 105 (205). Therefore, it is possible to measure the light quantity distribution in the direction orthogonal to the scanning direction of the imaging sensor 105 (205), which is condensed in the scanning direction of the imaging sensor 105 (205). By monitoring the light quantity distribution in this direction, it is possible to measure the temporal change of the light quantity distribution.

FIG. 7 is a flow chart showing main steps of the inspection method according to the first embodiment. 7, the inspection method in Embodiment 1 includes a calibration step (S102), a correction reference value setting step (S104), a scanning step (S110), a light amount distribution measuring step (S112), and a correction step. (S120), a reference image creation step (S130), and a comparison step (S140).

As the calibration step (S102), the imaging sensor 105 (205) is calibrated. For example, in the case of a sensor with 256 gradations, the sensitivity should be adjusted so that the gradation level when capturing an image of a white pattern is, for example, about 200 gradations. The light amount distribution of the illumination light at that time is measured by the line sensor 194 (294). The measured light amount distribution data is stored in the memory 196 (296) and output to the correction circuit 117 (207). The light amount distribution obtained here becomes the reference value at each position.

As the correction reference value setting step (S104), the correction circuit 117 (207) inputs the data of the light amount distribution measured by the line sensor 194 (294) when calibrated in the calibration step (S102), and corrects it. It is set as the reference value S0 for use.

In the scanning step (S110), the optical image acquisition mechanism 150 uses the calibrated imaging sensor 105 (205) to split the transmitted inspection light from the beam splitter 190 for inspection illumination light (second An image of the substrate 101 obtained by irradiating the substrate 101 on which the pattern is formed with the light of No. 1) is captured, and the captured optical image data is output. For this purpose, first, the optical image acquisition mechanism 150 scans the inspection stripe 20 with a laser beam (inspection light), and the imaging sensor 105 (205) captures a stripe area image for each inspection stripe 20 . Specifically, it operates as follows. The XYθ table 102 is moved to a position where the target inspection stripe 20 can be imaged. In the transmission inspection, light transmitted through the substrate 101 is imaged as an optical image on the imaging sensor 105 and enters. Alternatively/and in reflection inspection, reflected light from the substrate 101 is imaged as an optical image on the imaging sensor 205 and enters.

The pattern image formed on the imaging sensor 105 is photoelectrically converted by each photosensor element of the imaging sensor 105 and further A/D (analog/digital) converted by the sensor circuit 106 . At that time, the sensor circuit 106 converts the integrated outputs of the plurality of photosensor elements arranged in the scanning direction into gradation values and outputs the gradation values to the correction circuit 117 .

Alternatively/and the pattern image formed on the image sensor 205 is photoelectrically converted by each photosensor element of the image sensor 205 and further A/D (analog-digital) converted by the sensor circuit 206 . At this time, the sensor circuit 206 converts the integrated outputs of the plurality of photosensor elements arranged in the scanning direction into gradation values and outputs the gradation values to the correction circuit 217 .

In the light amount distribution measuring step (S112), in parallel with the imaging by the imaging sensor 105, the line sensor 194 receives the light (second light) split by the beam splitter 190 for measurement out of the transmitted inspection light. to measure the temporal change in the light amount distribution in the direction orthogonal to the accumulation direction of the imaging sensor 105 . In other words, the line sensor 194 simultaneously measures the illumination light used for imaging by the imaging sensor 105 . The measured light amount distribution data is stored and accumulated in the memory 196 .

Alternatively/and in parallel with the imaging by the imaging sensor 205, the line sensor 294 receives the light (second light) split by the beam splitter 290 for measurement out of the reflected inspection light, and outputs it to the imaging sensor 205. A temporal change in the light quantity distribution in the direction orthogonal to the accumulation direction is measured. In other words, the line sensor 294 simultaneously measures the illumination light used for imaging by the imaging sensor 205 . The measured light amount distribution data is stored and accumulated in the memory 296 .

In the correction step (S120), the correction circuit 117 (207) (correction unit) corrects the image data output from the imaging sensor 105 (205) according to the position of the light amount distribution using the value of the light amount distribution that changes with time. to correct.

FIG. 8 is a diagram for explaining a correction method according to the first embodiment. A plurality of photosensor elements arranged in the scanning direction arranged in the imaging sensor 105 (205) pick up images of the same pixels while shifting the time according to the movement of the XYθ table 102 . As shown in FIG. 8, the image sensor 105 (205) integrates and outputs data of the same pixel captured at different times by a plurality of photosensor elements arranged in the scanning direction. For example, N photosensor elements are arranged in the scanning direction. An image storage time (light receiving time, exposure time) per pixel is set for each photosensor element according to the moving speed of the XYθ table 102 . Therefore, for the time t calculated by the number of photosensor elements to be accumulated×image accumulation time, each pixel continues to receive illumination light for image acquisition. In addition, each photosensor element shifts the area of pixels on the substrate 101 to be imaged in the scanning direction for each image accumulation time.

On the other hand, the line sensor 194 (294) performs multiple measurements during the time t obtained by multiplying the number of photo sensor elements in the scanning direction used for imaging each pixel by the image accumulation time. For example, 1024 measurements are made. Data of the light amount distribution for each measurement is stored in the memory 196 (296).

The correction circuit 117 (207) corrects the image data of the target pixel using the value of the light quantity distribution measured during the time period when the image of the target pixel on the substrate 101 was captured by the imaging sensor 105 (205). Specifically, it operates as follows. The correction circuit 117 (207) reads out, from the memory 196 (296), data of a plurality of light amount distributions measured during the time period when the image of the target pixel on the substrate 101 was captured by the image sensor 105 (205). Then, the correction circuit 117 (207) calculates the moving average value S by totaling the value k at the corresponding position in the plurality of read light intensity distributions for each position of the light intensity distribution and dividing by the number of times of measurement n.

Next, the correction circuit 117 (207) corrects the image data of the target pixel using the moving average value of each value of the light quantity distribution measured multiple times during the time period when the image of the target pixel was captured. Specifically, the correction circuit 117 (207) corrects by multiplying the gradation value d of the target pixel calculated by the sensor circuit 106 (206) by a ratio obtained by dividing the reference value S0 by the moving average value S. In other words, correction is performed by multiplying the ratio of the amount of illumination light during calibration to the amount of light during actual imaging.

When the target pixel shifts in the scanning direction, the time period in which the image of the target pixel is captured also shifts while partially overlapping. Due to the deviation of the time zone, the plurality of light amount distribution data read from the memory 196 (296) are also different. Therefore, sets of a plurality of data sets of light amount distributions, which are the basis for correcting each pixel arranged in the scanning direction, also differ depending on the imaging time period. Therefore, since the moving average value S also becomes a different value, it is possible to perform correction according to the temporal change of the light intensity distribution.

At this time, the correction circuit 117 (207) corrects the image data at the target position using the value of the position of the light amount distribution corresponding to the target position of the image data in terms of the mask surface.

FIG. 9 is a diagram for explaining the relationship between magnification and image size at each position according to the first embodiment. The optical system may have different magnifications when forming an image. For example, as shown in FIG. 9, when the imaging region size of the mask substrate surface (magnification 1) is L1, the magnification is M1 on the imaging sensor 105 (205). In this case, the size of the image of the imaging region of the mask substrate surface imaged by the imaging sensor 105 (205) is the size obtained by multiplying L1 by the magnification M1. On the other hand, the magnification is M2 on the line sensor 194 (294). In this case, the size of the image of the imaging area of the mask substrate surface measured on the line sensor 194 (294) is the size obtained by multiplying L1 by the magnification M2. In this way, the size of the image L1 on the mask substrate is different on the imaging sensor 105 (205) and on the line sensor 194 (294). Therefore, each position of the light amount distribution measured by the line sensor 194 (294) does not necessarily correspond to the position in the direction orthogonal to the scanning direction of the image captured by the imaging sensor 105 (205). Corresponding positions differ depending on the difference in magnification. Therefore, a reference is required for the relationship between the position on the line sensor 194 (294) and the position on the imaging sensor 105 (205). Here, the position on the line sensor 194 (294) and the position on the imaging sensor 105 (205) are associated in terms of the mask surface.

FIG. 10 is a diagram for explaining the relationship between each position of the light amount distribution and the position of the image of the TDI sensor according to the first embodiment. The example of FIG. 10 shows the case where the image of the area from the position Y=Y0 to the position Y1 on the mask substrate is captured. In the example of FIG. 10, n3 photosensor elements are arranged in the line sensor 194 (294). On the other hand, in the imaging sensor 105 (205), m photosensor elements are arranged in the Y direction orthogonal to the scanning direction. For example, m is several times larger than n3. However, in terms of mask surface conversion, the position of Y=Y0 on the mask substrate corresponds to the n1-th pixel photosensor in the line sensor 194 (294). In this case, the image sensor 105 (205) corresponds to the first pixel photosensor. The position of Y=Y1 on the mask substrate corresponds to the n2-th pixel photosensor in the line sensor 194 (294). In this case, the image sensor 105 (205) corresponds to the m-th pixel photosensor. In this way, each position on the mask substrate may be assigned to each position on the line sensor 194 (294) and the imaging sensor 105 (205) in terms of the mask surface. In other words, the position on the line sensor 194 (294) and the position on the imaging sensor 105 (205) are associated in terms of the mask surface. If there is no data corresponding to the target position on the imaging sensor 105 (205) in the light amount distribution data, for example, a value obtained by linear interpolation using data of positions before and after the target position may be used. The longitudinal size of the line sensor 194 (294) should be equal to or greater than the longitudinal size of the imaging sensor 105 (205) in terms of the mask surface.

As described above, by using the light amount distribution in the direction perpendicular to the scanning direction, it is possible to perform correction according to the difference in the light amount in the direction perpendicular to the scanning direction. Furthermore, temporal changes in the amount of light in the scanning direction can be averaged by using a moving average value.

The gradation value (image data) of each pixel corrected by the correction circuit 117 (207) is output to the stripe pattern memory 123 (223).

Then, the pixel value data of the inspection stripe 20 to be measured is stored in the stripe pattern memory 123 (223). The measurement data (pixel data) is, for example, 8-bit unsigned data, and expresses the brightness gradation (light amount) of each pixel. The pixel value data of the inspection stripe 20 is output to the comparison circuit 108 .

As a reference image creation step (S130), the reference image creation circuit 112 creates a reference image that serves as a reference using figure pattern data (design data). The reference image is created for each inspection stripe 20 in parallel with the scanning operation of the inspection stripe 20 . Specifically, it operates as follows. The reference image generating circuit 112 inputs graphic pattern data (design data) for each frame region 30 of the target inspection stripe 20, and converts each graphic pattern defined in the graphic pattern data into binary or multi-valued image data. Convert to

The figures defined in the figure pattern data are, for example, rectangles and triangles as basic figures. The figure data defining the shape, size, position, etc. of each pattern figure is stored with information such as figure code, which is an identifier for each pattern.

When the design pattern data as such graphic data is input to the reference image generating circuit 112, it is developed into data for each graphic, and the graphic code, graphic dimensions, etc. indicating the graphic shape of the graphic data are interpreted. Then, it develops into binary or multi-valued design pattern image data as a pattern to be arranged in a grid of a predetermined quantization size as a unit, and outputs the data. In other words, the design data is read, and the occupancy rate of the figure in the design pattern is calculated for each square obtained by virtually dividing the frame area into squares having a predetermined size as a unit, and n-bit occupancy rate data ( design image data). For example, it is preferable to set one square as one pixel. Assuming that one pixel has a resolution of 1/2 ⁸ (=1/256), a small area of 1/256 is allocated for the area of the figure arranged in the pixel, and the occupancy rate in the pixel is reduced. Calculate. Then, it is created as 8-bit occupation rate data. Such squares (inspection pixels) may be aligned with the pixels of the measurement data.

Next, the reference image generation circuit 112 performs filter processing on the design image data of the design pattern, which is image data of the figure, using a filter function.

FIG. 11 is a diagram for explaining filter processing according to the first embodiment. Pixel data of an optical image captured from the substrate 101 is in a state in which a filter is applied due to the resolution characteristics of the optical system used for imaging, in other words, in a continuously changing analog state. In addition, the image intensity (gradation value) is different from the developed image (design image) of digital values. On the other hand, in the graphic pattern data, as described above, the graphic pattern data is defined by the graphic code or the like, so the image intensity (gradation value) may be a digital value in the developed design image. Therefore, the reference image creation circuit 112 performs image processing (filtering) on the developed image to create a reference image that is closer to the optical image. As a result, the design image data, which is image data on the design side in which the image intensity (gradation value) is a digital value, can be matched with the image generation characteristics of the measurement data (optical image). The created reference image is output to the comparison circuit 108 .

FIG. 12 is a diagram showing an example of an internal configuration of a comparison circuit according to Embodiment 1. FIG. 12, storage devices 70, 72, 76 such as magnetic disk devices, a frame image forming section 74, an alignment section 78, and a comparison processing section 79 are arranged in the comparison circuit 108. FIG. A series of "sections" such as the frame image creation section 74, the alignment section 78, and the comparison processing section 79 have processing circuits. Such processing circuits include electrical circuits, computers, processors, circuit boards, quantum circuits, or semiconductor devices. Also, each of the "-units" may use a common processing circuit (same processing circuit). Alternatively, different processing circuits (separate processing circuits) may be used. Input data necessary for the frame image creating section 74, the alignment section 78, and the comparison processing section 79 or the results of calculations are stored in the memory (not shown) in the comparison circuit 108 or the memory 111 each time.

The stripe data (stripe area image) input to the comparison circuit 108 is stored in the storage device 70 . The reference image data input to the comparison circuit 108 is stored in the storage device 72 .

As a comparison step (S140), the comparison circuit 108 (an example of a comparison unit) compares the image to be inspected using the corrected image data with a predetermined image. Specifically, it operates as follows.

In the comparison circuit 108, first, the frame image generator 74 generates a plurality of frame images 31 obtained by dividing the stripe area image (optical image) by a predetermined width. Specifically, as shown in FIG. 2, the stripe area image is divided into frame images of a plurality of rectangular frame areas 30 . For example, it is divided into a size of 512×512 pixels. Data for each frame area 30 is stored in the storage device 76 .

Next, the positioning unit 78 reads the corresponding frame image 31 and the corresponding reference image from the storage devices 72 and 76 for each frame region 30, and aligns the frame image 31 and the corresponding reference image with a predetermined algorithm. are aligned. For example, alignment is performed using the method of least squares.

Then, the comparison processing section 79 (another example of the comparison section) compares the frame image 31 with the reference image corresponding to the frame image 31 . For example, each pixel is compared. Here, both are compared for each pixel according to a predetermined determination condition, and the presence or absence of a defect such as a shape defect is determined. As a judgment condition, for example, both are compared for each pixel according to a predetermined algorithm to judge the presence or absence of a defect. For example, a difference value between pixel values of both images is calculated for each pixel, and if the difference value is greater than a threshold value Th, it is determined as a defect. The comparison result may be output to, for example, the magnetic disk device 109, a pattern monitor (not shown), or a printer (not shown).

In the above example, the case of die-to-database inspection was explained, but die-to-die inspection may also be used. In such a case, the comparison circuit 108 refers to the frame image (optical image) of the die 2 acquired for one of the frame regions for which the die-to-die inspection is performed among the plurality of frame regions 30. (reference image). First, the alignment unit 78 reads out the frame image 31 of the corresponding die 1 and the frame image of the die 2 from the storage device 76 for each frame region 30 in which the die-to-die inspection is performed. The frame image 31 and the frame image of the die 2 are aligned. For example, alignment is performed using the method of least squares. Then, the comparison processing unit 79 (comparing unit) compares the corresponding frame image 31 of the die 1 and the corresponding frame image of the die 2 pixel by pixel in each frame region 30 where the die-to-die inspection is performed.

As described above, according to the first embodiment, it is possible to measure changes in distribution unevenness of illumination light. Therefore, the change according to the position of the distribution unevenness can be corrected. As a result, the reproducibility of the device can be improved. Further, in the Koehler illumination, ideally, the in-plane uniformity is obtained. Therefore, even if distribution unevenness occurs, the difference in the unevenness is small. In Embodiment 1, for example, Koehler illumination light is created by the Koehler illumination optical system 300 . In Embodiment 1, it is possible to correct a change according to the position of the fine distribution unevenness caused by Koehler illumination.

In the above description, those described as "-circuit" have a processing circuit. Such processing circuits include, for example, electrical circuits, computers, processors, circuit boards, quantum circuits, or semiconductor devices. Each "-circuit" may use a common processing circuit (the same processing circuit) or may use different processing circuits (separate processing circuits). When using a program, the program is recorded on a recording medium such as a magnetic disk device, magnetic tape device, FD, or ROM (Read Only Memory).

The embodiments have been described above with reference to specific examples. However, the invention is not limited to these specific examples.

In addition, descriptions of parts that are not directly necessary for the explanation of the present invention, such as the device configuration and control method, are omitted, but the required device configuration and control method can be appropriately selected and used. For example, although the description of the configuration of the control unit that controls the inspection apparatus 100 has been omitted, it goes without saying that the required configuration of the control unit is appropriately selected and used.

In addition, all pattern inspection apparatuses and pattern inspection methods that have the elements of the present invention and whose designs can be modified as appropriate by those skilled in the art are included in the scope of the present invention.

10 inspection area 11 illumination light 20 inspection stripe 30 frame area 31 projection lens 32 beam splitters 70, 72, 76 storage device 74 frame image creating unit 78 alignment unit 79 comparison processing unit 100 inspection device 101 mask substrate 102 XYθ table 103 light source 104 Enlargement optical system 105, 205 Image sensor 106, 206 Sensor circuit 107 Position circuit 108 Comparison circuit 109 Magnetic disk unit 110 Control computer 111 Memory 112 Reference image creation circuit 114 Table control circuit 115 Drive mechanism 117, 217 Correction circuit 120 Bus 123, 223 Stripe pattern memory 134 Mirror 138 Mirror 150 Optical image acquisition mechanism 160 Control system circuit 176 Beam splitters 171, 271 Lens 170 Transmission illumination optical system 177 Mirrors 178, 278 Imaging optical system 190, 290 Beam splitter 192 Cylinder lenses 194, 294 Line sensor 196, 296 memory 270 reflected illumination optical system 300 Koehler illumination optical system 302 beam expander 304 division lens 306 rotating phase plate 308 collimator lens 310 illumination slit

Claims (6)

a beam splitter that splits the inspection light;
a time delay integration (TDI) sensor that captures an image of the mask substrate obtained by irradiating the mask substrate on which a pattern is formed with a first light branched from the inspection light and outputs image data;
a line sensor that receives the second light branched from the inspection light and measures a temporal change in the light amount distribution in a direction orthogonal to the accumulation direction of the TDI sensor;
a correction unit that corrects the image data output from the TDI sensor according to the position of the light intensity distribution using the value of the light intensity distribution that changes over time;
a comparison unit that compares the image to be inspected using the corrected image data with a predetermined image;
A mask inspection apparatus comprising: 2. The mask inspection according to claim 1, wherein the correcting unit corrects the image data at the target position by using the value of the position of the light amount distribution corresponding to the target position of the image data in terms of the mask surface. Device. The correcting unit corrects the image data of the target pixel by using the value of the light amount distribution measured during the time period when the image of the target pixel on the mask substrate is captured by the TDI sensor. 3. The mask inspection apparatus according to item 1 or 2. 4. The mask inspection apparatus according to claim 1, wherein said inspection light is illuminated using Koehler illumination. The correction unit corrects the image data of the target pixel by using a moving average value of each value of the light amount distribution measured during the time period when the image of the target pixel on the mask substrate is captured by the TDI sensor. 5. The mask inspection apparatus according to any one of claims 1 to 4, characterized by: a step of branching the inspection light;
A time delay integration (TDI) sensor is used to pick up an image of the mask substrate obtained by irradiating the mask substrate on which the pattern is formed with the first light branched out of the inspection light, and output image data. and
a step of using a line sensor to receive the second light branched from the inspection light and measuring a temporal change in the light amount distribution in a direction perpendicular to the accumulation direction of the TDI sensor;
a step of correcting the image data output from the TDI sensor according to the position of the light intensity distribution using the value of the light intensity distribution that changes with time;
comparing the image under inspection using the corrected image data to a predetermined image and outputting the results;
A mask inspection method comprising:

Images