JP2012059984A - Mask inspection device and exposure mask manufacturing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a mask inspection device capable of detecting phase defect at the stage of mask blank that is a stage before depositing a absorber pattern.SOLUTION: A mask inspection device 100 includes: an illumination optical system that illuminates illumination light on the surface of mask blank in which a multilayer film for reflecting extreme ultraviolet (EUV) light is formed; sensors 105 and 305 that images an image of the same location as a location to which the light is reflected from the surface of the mask blank at a location in which defocus is performed with a different defocus mount; and a determination part 176 that determines the presence or absence of defect of the mask blank using first and second optical images at the same location on the mask blank surface imaged with the different defocus amount.

Description

  The present invention relates to a mask inspection apparatus and an exposure mask manufacturing apparatus, for example, an inspection apparatus for inspecting a mask blank defect of a mask used for extreme ultraviolet (EUV) exposure, and a further inspected mask blank. The present invention relates to an exposure mask manufacturing apparatus for drawing a pattern.

  Lithography technology, which is responsible for the progress of miniaturization of semiconductor devices, is an extremely important process for generating a pattern among semiconductor manufacturing processes. In recent years, with the high integration of LSI, circuit line widths required for semiconductor devices have been reduced year by year. In order to form a desired circuit pattern on these semiconductor devices, a highly accurate original pattern (also referred to as a reticle or a mask) is required.

  Here, with the miniaturization of semiconductor devices, it has been studied to make the exposure wavelength itself shorter than before. The light of 157 nm has been abandoned due to the limitation of the lens material for reduced transfer. Therefore, it is EUV light having a wavelength of 13.4 nm that is considered to be most likely at the present time. Since EUV light is transmitted and absorbed by many objects with light divided into a soft X-ray region, it is no longer possible to form a projection optical system. Therefore, a reflection optical system has been proposed for an exposure method using EUV light. As described above, in the EUV lithography, a reflection optical system including a multilayer mirror that reflects EUV light is used. An EUV mask for EUV exposure is interposed as a part of the optical system. Therefore, similarly, a reflective mask in which a multilayer film is formed on a substrate is used. A multilayer film in which molybdenum (Mo) and silicon (Si) layers are alternately stacked is used. An absorber pattern is formed on the mask blank on which the laminated film is formed to constitute a mask for EUV lithography (EUVL).

  Here, when the regularity of the thickness of each layer of the laminated film is lost, the phase of the reflected light is shifted. As a result, the wafer is exposed as a phase defect. In a conventional transmissive mask blank for optical lithography, irregularities of about several nm on the surface can be ignored. However, in the EUV mask, even when an abnormality with a height of only a few nanometers occurs on the mask blank on which such a multilayer film is formed, the EUV reflected light causes a large phase change due to the abnormal height, and the absorber pattern Defects are produced during transfer onto the wafer. Therefore, the EUV mask has a large qualitative difference in defect transfer compared to the conventional transmission mask, and it is necessary to avoid the generation of a mask blank defect that gives a phase difference.

  Therefore, it is necessary to detect the phase defect at the stage of the mask blank before depositing the absorber pattern. As an inspection method, a defect is detected by irradiating a mask blank with laser light and detecting foreign matter from light that is diffusely reflected (for example, see Patent Document 1), or using detection light having the same wavelength as EUV light used for exposure. A method (for example, see Patent Document 2) has been studied. However, in the inspection method in which foreign matter is detected from light that is irregularly reflected using laser light, the height of the phase defect to be detected is two or more orders of magnitude smaller than the inspection wavelength, which makes it difficult to detect minute phase defects. There's a problem. On the other hand, in the inspection method using EUV light, it is difficult to handle the inspection optical system in the vacuum chamber that employs the EUV light source and the multilayer film reflection surface, and the damage to the optical mirror is not sufficiently small. .

  Furthermore, as a result of the inspection, it is difficult to completely eliminate the mask blank defects, and if all the manufactured masks are inspected and only those having zero defects or satisfying the specifications are selected, the EUV mask is very expensive. It will become something.

  Therefore, the phase defect on the multilayer mask is not transferred by moving the absorber pattern so that the defect is not transferred by the exposure process so that the phase defect is included in the region of the absorber pattern. The technique of doing is examined (for example, refer patent document 3).

JP 2001-174415 A JP 2003-114200 A JP 2004-193269 A

  As described above, it is necessary to detect the phase defect at the stage of the mask blank before depositing the absorber pattern. However, a sufficient method for solving such a problem has not been established.

  Therefore, an object of the present invention is to provide an apparatus capable of overcoming such a problem and detecting a phase defect at a stage of a mask blank before depositing an absorber pattern.

A mask inspection apparatus according to an aspect of the present invention includes:
An illumination optical system that illuminates illumination light on the surface of a mask blank on which a multilayer film that reflects extreme ultraviolet (EUV) light is formed;
First and second sensors that capture images at the same position reflected from the mask blank surface at different defocus amounts;
A determination unit that determines the presence or absence of a defect in the mask blank using the first and second optical images at the same position on the surface of the mask blank imaged with different defocus amounts;
It is provided with.

  With this configuration, it is possible to detect a mask blank defect as described later.

Further, the image processing apparatus further includes a difference calculation unit that calculates a difference between the first and second optical images for each pixel,
It is preferable that the determination unit determines a defect for each pixel when the difference exceeds a threshold value.

The first sensor captures the first optical image on the front side of the imaging point of the first optical image,
The second sensor preferably captures the second optical image behind the imaging point of the second optical image.

  Further, it is preferable to use deep ultraviolet (DUV) light having a wavelength of 180 to 210 nm as illumination light.

An exposure mask manufacturing apparatus according to an aspect of the present invention includes:
An illumination optical system that illuminates illumination light on the surface of a mask blank on which a multilayer film that reflects extreme ultraviolet (EUV) light is formed;
First and second sensors that capture images at the same position reflected from the mask blank surface at different defocus amounts;
A determination unit that determines the presence or absence of a defect in the mask blank using the first and second optical images at the same position on the surface of the mask blank imaged with different defocus amounts;
A drawing unit that draws a pattern on a mask blank on which another film is further formed using a charged particle beam so that a position determined to have a defect is removed from the EUV light reflection surface when performing EUV exposure;
It is provided with.

  According to one aspect of the present invention, phase defects can be detected at the stage of the mask blank before the absorber pattern is deposited. Further, according to another aspect of the present invention, when the wafer is exposed with the manufactured mask, the phase defect existing in the mask can be prevented from being transferred onto the wafer.

1 is a conceptual diagram showing a configuration of a mask inspection apparatus in a first embodiment. 5 is a diagram showing an example of an EUV mask in the first embodiment. FIG. It is a figure which shows an example of the mask blank and EUV mask in which the defect in Embodiment 1 exists. It is a figure which shows an example of the light intensity of the position on the mask blank in Embodiment 1, and the imaged image. An example of the position on the mask blank in Embodiment 1 and the light intensity of the imaged image is shown. 3 is a conceptual diagram for explaining an optical image acquisition method according to Embodiment 1. FIG. 6 is a conceptual diagram for explaining a method of acquiring images of two regions in the first embodiment. FIG. FIG. 6 is a time chart for explaining the flow of reading and reading measurement data into the memory in the first embodiment. FIG. 3 is a flowchart showing main steps of the mask inspection method in the first embodiment. It is a figure which shows an example of the positional relationship of the defect position and absorber pattern in Embodiment 2. FIG. It is a conceptual diagram which shows the structure of the mask manufacturing apparatus in Embodiment 2. FIG. 10 is a conceptual diagram illustrating a configuration of a drawing unit in a mask manufacturing apparatus according to a second embodiment. It is a top surface conceptual diagram which shows an example of the EUV mask in which the defect in Embodiment 2 exists. FIG. 14 is a conceptual cross-sectional view showing a cross section of the EUV mask of FIG. 13. It is a conceptual diagram which shows an example in the case of drawing a pattern on the mask blank in which the defect in Embodiment 2 exists.

Embodiment 1 FIG.
FIG. 1 is a conceptual diagram showing the configuration of the mask inspection apparatus in the first embodiment. In FIG. 1, a mask inspection apparatus 100 includes an optical image acquisition unit and a control unit. The optical image acquisition unit includes a light source 103, lenses 122 and 124, a slit plate 123, a beam splitter 126, an objective lens 104, an XYθ table 102, a reflection mirror 128, lenses 132 and 332, and imaging sensors 105 and 305. Yes. The lenses 122 and 124, the slit plate 123, the beam splitter 126, and the objective lens 104 constitute an illumination optical system. On the XYθ table 102, a mask blank 101 on which a multilayer film reflecting extreme ultraviolet (EUV) light is formed is placed with the multilayer film surface down. As the mask blank 101, a mask blank at a stage where a multilayer film is formed and still before the absorber film is formed is used. In the first embodiment, the mask blank 101 at the stage where such a multilayer film is formed is an inspection target. The XYθ table 102 is driven by an X-axis motor, a Y-axis motor, and a θ-axis motor.

  The control unit includes sensor circuits 106 and 306, memories 107 and 307, a comparison circuit 108, a control computer 110, a storage device 109 such as a magnetic disk device, a table control circuit 111, a monitor 118, and a printer 119. The memories 107 and 307, the comparison circuit 108, the control computer 110, the storage device 109 such as a magnetic disk device, the table control circuit 111, the monitor 118, and the printer 119 are connected to each other via a bus 120. In the comparison circuit 108, an alignment processing unit 172, a difference calculation unit 174, and a determination unit 176 are arranged. Each function of the alignment processing unit 172, the difference calculation unit 174, and the determination unit 176 may be configured by software such as a program for causing a computer to execute. Or you may comprise by hardware, such as an electric equipment or an electronic device. Alternatively, a combination of software and hardware may be used. Alternatively, a combination of firmware and hardware may be used. The table control circuit 111 moves the XYθ table 102 by controlling the X-axis motor, the Y-axis motor, and the θ-axis motor.

  The light source 103 generates deep ultraviolet (DUV) light having a wavelength of 180 to 210 nm. Here, DUV light having a wavelength of 199 nm is used as illumination light. The DUV light generated from the light source 103 illuminates the surface of the mask blank 101 on which the multilayer film is formed by the illumination optical system. Specifically, after the DUV light generated from the light source 103 passes through the lens 122, it is narrowed down to two light beams by two slit openings formed in the slit plate 123. As described above, the two light beams 10 that have passed through the two slit openings pass through the lens 124, have their optical paths bent by the beam splitter 126, are condensed by the objective lens 104, and are irradiated onto the surface of the mask blank 101. . The two light beams illuminate the adjacent regions 11 and 12 in the region 1LA on the surface of the mask blank 101. The light beams 13 of the two images in the regions 11 and 12 reflected from the surface of the mask blank 101 pass through the objective lens 104 and the beam splitter 126. The light beams 13 of the two images that have passed through the beam splitter 126 are condensed on the projection area PRA. In this manner, the regions 11 and 12 on the mask blank 101 are formed by separating the images 14 and 15 in the projection region PRA.

The light beam that forms the image 14 (first optical image) has its optical path bent by the reflecting mirror 128 disposed immediately below the projection area PRA, passes through the lens 332, and is disposed in front of the imaging point 17 of the light beam. The image is picked up by the sensor 305. That is, the image 14 is formed on the light receiving surface of the sensor 305. The light beam forming the image 15 (second optical image) passes through the lens 132 and is picked up by the sensor 105 disposed on the rear side of the image forming point 18 of the light beam. That is, the image 15 is formed on the light receiving surface of the sensor 105. As described above, the imaging sensors 105 and 305 capture the two images reflected from the surface of the mask blank 101 at positions defocused with different defocus amounts. In the example of FIG. 1, the sensor 305 is arranged at a position where the optical path length is shorter by XL1 than the imaging point 17 of the lens 332, and the sensor 105 is arranged at a position where the optical path length is longer by XL2 than the imaging point 18 of the lens 132. Shows the case. With this arrangement, the two sensors 105 and 305 can collect the inspection images of the adjacent regions 11 and 12 on the mask blank in a defocused state with different signs. These defocus amounts are XL1 / (magnification) ² and XL2 / (magnification) ² when converted to defocus amounts on the mask blank surface.

  The image formed on the sensor 105 is photoelectrically converted by the sensor 105 and further A / D (analog / digital) converted by the sensor circuit 106. The sensor 105 is provided with a line sensor (or a two-dimensional area sensor) such as a TDI (time delay integrator) sensor. Similarly, an image formed on the sensor 305 is photoelectrically converted by the sensor 305 and further A / D converted by the sensor circuit 306. The sensor 305 is provided with a line sensor (or a two-dimensional area sensor) such as a TDI sensor. By continuously moving the XYθ table 102 serving as a stage, for example, in the X-axis direction, the two TDI sensors continuously capture images on the surface of the mask blank 101. Thereby, the area | regions 11 and 12 with which the mask blank 101 surface is illuminated are imaged, moving continuously. Therefore, the image at the same position on the surface of the mask blank 101 is picked up at the position defocused by the imaging sensors 105 and 305 with different defocus amounts with a time difference.

  FIG. 2 is a diagram showing an example of the EUV mask in the first embodiment. FIG. 2A shows an example of the EUV mask 301 viewed from the pattern surface, and has a device pattern area 2 having a pattern of a semiconductor integrated circuit device at the center. In addition, alignment mark areas MA1, MA2, MA3, MA4 including a mask alignment mark, a wafer alignment mark, and the like are arranged on the periphery of the EUV mask 301. FIG. 2A shows an example of a cross section of the EUV mask 301. A multilayer film ML is deposited on a substrate MS such as quartz glass or a low thermal expansion material, and a capping layer CAP is deposited thereon. On top of this, an absorber pattern ABS is provided via a buffer layer BUF. Further, a metal film CF for electrostatic chucking the mask is coated on the back surface side of the substrate MS. The mask blank 101 to be inspected in the first embodiment is in a stage where a multilayer film ML is deposited on a substrate MS such as quartz glass or a low thermal expansion material. Alternatively, the multilayer film ML and the capping layer CAP are deposited on the substrate MS.

  FIG. 3 is a diagram illustrating an example of a mask blank and an EUV mask in which defects exist in the first embodiment. In FIG. 3A, when the multilayer film ML is deposited on the substrate MS, the multilayer film ML is deposited on the substrate MS while the minute depression is present. An example in which PD has occurred will be described. When the buffer layer BUF and the absorber pattern ABS are formed while leaving this defect, for example, as shown in FIG. 3B, the phase defect PD is formed on the bottom surface (reflection surface) of the opening between the adjacent absorber patterns ABS. May remain. When this phase defect exists, the pattern projection image transferred onto the wafer is disturbed and a defect in the transfer pattern occurs.

  FIG. 4 is a diagram showing an example of the position on the mask blank and the light intensity of the captured image in the first embodiment. FIG. 4 shows a result obtained by simulation analysis of the light intensity distribution of the magnified image formed on the light receiving surfaces of the sensors 105 and 305 when the concave phase defect PD1 exists on the multilayer film ML of the mask blank 101. ing. The depth of the phase defect was 2.2 nm, the full width at half maximum was 60 nm, and the numerical aperture NA of the objective lens was 0.75. Here, it is obtained when the position of the mask blank 101 is shifted from the predetermined position along the optical axis direction of the objective lens 104, that is, when the defocus amount is set to -400 nm, -200 nm, 0 nm, +200 nm, +400 nm. Enlarged projection image light intensity distributions are shown by curves 1-1, 1-2, 1-3, 1-4, and 1-5, respectively. From this figure, it can be seen that when the defocus amount is 0, the projected image light intensity corresponding to the defective portion is slightly darker than the surrounding light intensity that is not the defective portion. At the same time, with respect to negative defocus such as −200 nm and −400 nm, the light intensity at the defect portion is higher than the light intensity at the peripheral portion. On the other hand, it can be seen that for positive defocus such as +200 nm and +400 nm, the light intensity at the defect portion is lower than the light intensity at the peripheral portion. In addition, it can be seen that the light intensity greatly changes in the defective portion by defocusing compared to the peripheral portion.

  FIG. 5 shows an example of the position on the mask blank and the light intensity of the captured image in the first embodiment. In FIG. 5, when the convex phase defect PD2 exists on the multilayer film ML of the mask blank 101, the result obtained by simulation analysis of the light intensity distribution of the magnified image formed on the light receiving surfaces of the sensors 105 and 305 is shown. Show. Here, the height of the convex phase defect PD2 is 2.2 nm, the full width at half maximum is 60 nm, and the numerical aperture NA of the objective lens is 0.75. Here, the curves 2-1, 2-2, 2-3, 2-4, 2-5 representing the enlarged projected image light intensity distribution are defocus amounts of +400 nm, +200 nm, 0 nm, −200 nm, − Corresponds to 400 nm. As shown in FIG. 5, when the defocus amount is set to 0, the projected image light intensity corresponding to the defective part becomes darker than the surrounding light intensity, and the light intensity corresponding to the defective part depends on the defocusing of light and dark. It can be seen that the sign is inverted compared to the case of the concave phase defect. Here again, it can be seen that the light intensity greatly changes in the defective portion by defocusing compared to the peripheral portion.

  In the first embodiment, defect detection is performed using the above facts. In other words, in the first embodiment, a defect of the mask blank 101 is detected using two optical images at the same position on the surface of the mask blank 101 captured with different defocus amounts.

  FIG. 6 is a conceptual diagram for explaining an optical image acquisition method according to the first embodiment. In FIG. 6, the inspection region 21 on the mask blank 101 is virtually divided into a plurality of strip-shaped inspection stripes 22 with a predetermined width in the y direction, for example. The width of the inspection stripe 22 is set to a line width (pixel width) that can be imaged at one time by the sensors 105 and 305. For example, it is preferable that the length is set to 2048 pixels. Here, as an example, in each of the sensors 105 and 305, 2048 pixels of light receiving elements are arranged in the pixel width direction (line width direction), and 512 light receiving elements are arranged in the integration direction (pixel column direction). TDI sensor is used. This TDI sensor transfers the charges received by the light receiving elements in each column in the moving direction of the XYθ table 102 to the rear column one by one in the moving direction of the XYθ table 102, so as to store the charges by the number of storage stages. Accumulated and output. As the XYθ table 102 moves in the −x direction, for example, the sensors 105 and 305 continuously capture images in the x direction indicated by the arrow 23. At that time, the reflected image from the mask blank 101 at the same time is divided into two regions IMA1 and IMA2. The sensor 105 captures an image of the area IMA2, for example, and the sensor 305 captures an image of the area IMA1, for example, at the same time. Images of two adjacent areas IMA1 and IMA2 are captured as the intensities (or pixel values) of the plurality of pixels 24, respectively.

  FIG. 7 is a conceptual diagram for explaining a method of acquiring images of two regions in the first embodiment. In FIG. 7, it is a figure which shows the positional relationship of the area | regions IMA1 and IMA2 taken in by the sensors 105 and 305 at various times. A state indicated by T1 is a state immediately before the scanning of the XYθ table 102 starts and the front end of the inspection stripe 22 approaches the area IMA1 to be captured by the sensor 305. The state indicated by T <b> 2 is a state immediately before the scanning of the XYθ table 102 advances and the tip of the inspection stripe 22 reaches the area IMA <b> 2 where the sensor 105 takes in. A state indicated by T3 is a state in which the scanning of the XYθ table 102 is completed, and indicates a state in which the final end of the inspection stripe 22 exceeds the area IMA2 to be captured by the sensor 105.

  FIG. 8 is a time chart for explaining the flow of reading and reading measurement data into the memory in the first embodiment. Here, processing of measurement data of one inspection stripe 22 will be described. Scanning of the XYθ table 102 on which the mask blank 101 is placed is started at time t1. When the light receiving elements in the first column of the sensor 305 reach the inspection stripe 22 at time t2, the positional relationship between the position of the XYθ table 102 and the image captured by the sensor 305 is stored, and at the same time, capturing of inspection image data is started. Data is sequentially stored in 307. When the light receiving elements in the first column of the sensor 105 reach the inspection stripe 22 at time t3, the positional relationship between the position of the XYθ table 102 and the image captured by the sensor 105 is stored, and at the same time, the acquisition of inspection image data is started. Data is sequentially stored in 107. At time t4, the capture of inspection image data by the sensor 305 is terminated, and at time t5, capture of inspection image data by the sensor 105 is terminated. At time t6, the scanning of the mask blank 101 is finished to prepare for the transition to the next inspection stripe.

  Here, during the time t3 to the time t5, the inspection image data captured by the sensor 105 is taken into the memory 107, and at the same time, the measurement data captured by the sensor 305 recorded in advance is read out from the memory 307 sequentially. The inspection image data at the same location on the mask blank 101 is compared. Since both images are data collected in a defocused state with different signs, if there is a phase defect, as shown in FIGS. 4 and 5, the inspection image intensity of the defective portion is one signal that is dark and the other is a bright signal. It becomes.

  Therefore, the comparison circuit 108 sequentially reads out the respective measurement data (first and second optical images) imaged by the sensors 105 and 305 from the memories 107 and 307, and on the mask blank 101 imaged with a time difference. Compare images at the same position. First, the alignment processing unit 172 performs alignment (alignment) between images at the same position. In this case, it is preferable that the subpixel unit is further combined after the pixel unit. For example, alignment is performed using a least square method.

  Next, the difference calculation unit 174 calculates the difference between both images for each pixel. In other words, the difference (absolute value of the difference) is calculated for pixels at the same position. The determination unit 176 determines a defect for each pixel when the difference (difference signal intensity) exceeds a threshold value. Thereby, a phase defect is detected. As described above, the determination unit 176 determines the presence / absence of a defect in the mask blank 101 using the two optical images at the same position on the surface of the mask blank 101 captured with different defocus amounts. The determination result is stored in the storage device 109. Then, it is displayed on the monitor 118. Alternatively, it may be output to a paper medium by the printer 119.

  FIG. 9 is a flowchart showing main steps of the mask inspection method according to the first embodiment. First, in S (step) 101, the first inspection stripe is designated in the region to be inspected on the mask blank.

  Thereafter, in S102, the continuous movement of the stage (XYθ table 102) on which the mask blank 101 is placed is started, and the capture of the inspection image is started.

  In S103, when the light receiving element of the sensor 305 enters the region of the inspection stripe 22, the inspection image data by the DUV light is captured, and at the same time, the inspection image is captured into the memory 307.

  In S104, when the light receiving element of the sensor 105 enters the region of the inspection stripe 22 by continuous scanning of the mask blank 101, the inspection image data is captured and simultaneously the inspection image is captured into the memory 107. Further, the inspection images collected by the sensor 305 are sequentially read from the memory 307.

  In S105, a difference image is calculated for each corresponding pixel in the same region on the mask blank 101.

  In S106, the intensity of the difference image is compared with a predetermined threshold value, and if there is a pixel whose intensity of the difference image exceeds the threshold value, it is determined that a phase defect exists.

  If it is determined in S107 that the phase defect is present, defect information is displayed or recorded as necessary. When the threshold is exceeded over a plurality of pixels, the size and contour of the defect can be grasped. When calculating the difference image, it is possible to determine the unevenness of the defective portion from the magnitude relationship between the intensities of the inspection image collected by the sensor 305 and the inspection image collected by the sensor 105.

  In S108, it is determined whether all the processes on one stripe have been completed. If not completed, the process returns to S103 to continue the inspection. When all the processes on one stripe are completed, the process proceeds to S109.

  In S109, it is determined whether or not the defect inspection of all inspection areas on the mask blank 101 has been completed. If not completed, the process proceeds to S110.

  In S110, a new inspection stripe is designated, the mask blank 101 is positioned at the start position of the next stripe, and the process returns to step S102.

  As described above, S102 to S110 are repeated, and when the processing on all the stripes in the predetermined area on the mask is finished, the defect inspection is finished.

  Here, in the first embodiment, the method of phase defect inspection of the mask blank 101 is shown, but the present invention is not limited to this. Even in an EUV mask in which the absorber is partially formed, the same phase defect inspection can be performed in a region where the absorber pattern is removed and the multilayer film reflecting surface is exposed.

  In addition, an example in which the wavelength of the inspection light is 199 nm has been shown, but the inspection is performed not only in this wavelength region but also by calculating a difference image between inspection images in different defocus states to emphasize the phase defect detection signal. Can be performed.

  Further, the coordinates of the detected phase defect can be determined based on marks provided on the mask blank 101 as appropriate, or in the case of an EUV mask with an absorber, based on the position of a predetermined absorber pattern.

  In the first embodiment, two optical images at the same position on the surface of the mask blank 101 captured with different defocus amounts are used. However, the present invention is not limited to this. One image may be a focused image. That is, the defocus amount may be zero for one image. In addition, the defocus amounts of the two images are not limited to different signs, and may be the same sign depending on circumstances. What is necessary is just to set so that the difference of the pixel value of two images may become large compared with the position without a defect even if it is the same code | symbol.

  In the first embodiment, the mask blank 101 is illuminated after being divided into two illumination lights by the slit plate 123, but is not limited thereto. One illumination light may be irradiated to the mask blank 101, and the reflected light from the mask blank 101 may be branched in the middle using a beam splitter. Then, the two branched light beams may be imaged at defocus positions with different defocus amounts. The amount of light is halved by branching, but the amount of illumination light may be increased to such an extent that it can be inspected even if branched. By branching, images at the same position can be acquired at the same time.

  As described above, according to the first embodiment, the phase defect detection signal can be emphasized using the mask pattern defect inspection apparatus using the DUV light as the inspection light, and the phase defect remaining on the mask or the mask blank. Can be detected with high sensitivity. Therefore, the phase defect can be detected at the stage of the mask blank 101 before the absorber pattern is deposited. Furthermore, since DUV light is used as illumination light, handling of the inspection optical system is simplified compared to the case of using EUV light.

Embodiment 2. FIG.
In the first embodiment, the inspection apparatus and the inspection method for inspecting the defect of the mask blank 101 have been described. However, in the second embodiment, a configuration in which a pattern drawing for manufacturing an EUV mask is added will be described. To do.

  FIG. 10 is a diagram illustrating an example of a positional relationship between a defect position and an absorber pattern in the second embodiment. As shown in FIG. 10A, if the absorber pattern is arranged so that the absorber pattern ABS covers the phase defect PD remaining on the mask blank, the phase defect PD is not transferred to the wafer. . Further, when the defect PD on the mask blank is small so as not to affect the transfer by itself, if there is no absorber pattern in the vicinity thereof, it does not cause a variation in the dimension of the pattern to be projected and exposed. . Therefore, for example, as shown in FIG. 10B, the absorber pattern is positioned so that the absorber pattern ABS formed on the multilayer film surface 31 is sufficiently separated from the phase defect PD only when the defect PD is fine. Just do it.

  FIG. 11 is a conceptual diagram showing the configuration of the mask manufacturing apparatus in the second embodiment. In FIG. 11, the mask manufacturing apparatus 500 according to the second embodiment includes an inspection unit 300 and a drawing unit 400. The configuration of the inspection unit 300 is the same as that of the mask inspection apparatus 100 shown in FIG. The inspection method is the same as that of the first embodiment. Further, the contents not specifically described below are the same as those in the first embodiment. First, the inspection unit 300 performs a defect inspection of the mask blank 101. The inspection method is the same as in the first embodiment. In the inspected mask blank 101, an absorber film is formed on the already formed multilayer film (and capping layer), and a resist that can be exposed by an electron beam is applied thereon.

  FIG. 12 is a conceptual diagram illustrating a configuration of a drawing unit in the mask manufacturing apparatus according to the second embodiment. In FIG. 12, a drawing unit 400 is an example of a charged particle beam drawing apparatus. Here, in particular, an example of a variable shaping type electron beam drawing apparatus is shown. The drawing unit 400 includes a drawing processing unit 450 and a control unit 460. Then, the drawing unit 400 uses the electron beam 200 to draw a desired pattern on the mask blank 101 on which an absorber film is formed on a multilayer film (capping layer) and a resist is applied thereon.

  The drawing processing unit 450 includes an electronic lens barrel 402 and a drawing chamber 403. In the electron column 402, an electron gun 201, an illumination lens 202, a first aperture 203, a projection lens 204, a deflector 205, a second aperture 206, an objective lens 207, and a deflector 208 are arranged. In the drawing chamber 403, an XY stage 405 that is movably disposed is disposed. When drawing a pattern, a plurality of support pins 406 (an example of a holding unit) are arranged to be movable up and down on the XY stage 405, and a mask blank 101 to be drawn is placed on the support pins 406.

  The control unit 460 includes a computer unit 410, a control circuit 462, and storage devices 440 and 442 such as a magnetic disk device. The computer unit 410, the control circuit 462, and the storage devices 440 and 442 such as a magnetic disk device are connected to each other via a bus (not shown).

  The control circuit 462 is controlled by the drawing data processing unit 76, and the control circuit 462 controls and drives each device in the drawing processing unit 450.

  In the control computer unit 410, a defect coordinate / size acquisition unit 62, a partial pattern data / shift amount and direction calculation unit 64, a drawing data processing unit, and a memory 78 are arranged. The functions of the defect coordinate / size acquisition unit 62, the partial pattern data / shift amount and direction calculation unit 64, and the drawing data processing unit 76 may be configured by software such as a program that executes a computer. Or you may comprise by hardware, such as an electric equipment or an electronic device. Alternatively, a combination of software and hardware may be used. Alternatively, a combination of firmware and hardware may be used. Input information and calculation processing information processed by the functions of the defect coordinate / size acquisition unit 62, the partial pattern data / shift amount and direction calculation unit 64, and the drawing data processing unit 76 are stored in the memory 78 each time.

  First, the defect position and size obtained as a result of the inspection by the inspection unit 300 are stored in the storage device 440. For example, an identifier (ID) may be set for the mask blank 101 to be drawn and stored so as to correspond to the ID. The coordinates indicating the defect position may be used with reference to a mark such as a step pattern provided in advance on the mask blank 101, whereby the position coordinates of the defect can be accurately grasped. For example, the defect position is indicated by coordinates (x, y) of the center position of the defect with reference to the coordinates of the mark position. In addition, the defect size is preferably represented by a diameter (D) of a circle expanded to the maximum outer shape of the defect with the defect position coordinate as the center. Further, drawing data serving as layout pattern data is input from the outside and stored in the storage device 442.

  First, the defect coordinate / size acquisition unit 62 refers to the defect position and size stored in the storage device 440 and reads defect position information indicating the defect position of the mask blank 101 and defect size information indicating the defect size. .

  Next, the partial pattern data / shift amount and direction calculation unit 64 inputs partial pattern data corresponding to an area including at least a defect in the drawing data stored in the storage device 442. Further, the partial pattern data / shift amount and direction calculation unit 64 simultaneously calculates the shift amount and the shift direction by shifting the pattern layout so that the defect is hidden under the absorber pattern.

  FIG. 13 is a top conceptual view showing an example of an EUV mask having defects in the second embodiment. FIG. 14 is a conceptual cross-sectional view showing a cross section of the EUV mask of FIG. The EUV mask has a multilayer film 512 in which, for example, 40 layers of molybdenum (Mo) with a thickness of 2.9 nm and silicon (Si) with a thickness of 4.1 nm are alternately stacked on the glass substrate 510. Is formed. Then, a cap film 514 such as ruthenium (Ru) is formed on the entire surface of the multilayer film 512. The cap film 514 is exposed in the region that reflects EUV light. On the other hand, in the region that does not reflect EUV light, an absorber film 516 that absorbs EUV light and an antireflection film 518 are sequentially formed on the cap film 514. Here, as shown in FIGS. 13A and 14A, when the defect 540 of the multilayer film 512 is present in the region 542 where the absorber film 516 does not exist, the phase of the EUV light reflected as described above changes. It will shift. As a result, when the pattern is transferred onto the semiconductor wafer with the manufactured EUV mask, the position of the pattern is shifted. Therefore, in the second embodiment, as shown in FIGS. 13B and 14B, after patterning, the position of the defect 540 is within the region 544 where the absorber film 516 exists, as shown in FIG. Drawing is performed by shifting the pattern layout from the positions shown in a) and FIG. In other words, the anti-reflection film 518 and the absorber film 516 are drawn on the EUV mask blank 101 coated with the resist film by the drawing unit 400, the resist is developed, and the resist pattern formed by the resist film remaining after the development is used as a mask. The EUV mask blank 101 is patterned by removing the remaining resist film by ashing. An EUV mask is manufactured by such patterning. Then, after such patterning, the pattern layout is shifted when drawing by the drawing unit 400 so that the position of the defect 540 is within the region 544 where the absorber film 516 exists.

  FIG. 15 is a conceptual diagram showing an example when a pattern is drawn on a mask blank having a defect in the second embodiment. ALN-marks 52 and 54 are formed on the mask blank 101. The inspection unit 300 inspects the presence or absence of the defect 540 on the mask blank 101 and measures the defect position and the defect size with reference to the ALN-marks 52 and 54. As a result, the defect position and defect size information based on the positions of the ALN-marks 52 and 54 can be acquired in advance before drawing.

  For example, the example of FIG. 15 shows a case where a defect 540 exists in a partial area 56 of the drawing area 50 of the mask blank 101. If the pattern layout is used as it is, the defect 540 is located in the region 542 where the absorber film 516 does not exist, thereby causing a phase defect. On the other hand, in the second embodiment, the pattern for drawing is shifted so that the defect 540 is located in the region 544 where the absorber film 516 exists by shifting the pattern layout so that the partial region 56 becomes the partial region 58. Correct the data.

  The drawing data processing unit 76 reads the drawing data from the storage device 442 and performs a plurality of stages of data conversion processing in the parallel processing and the processing for calculating the shift amount and the shift direction described above, and the shot becomes a device-specific format. Generate data. Alternatively, processing may be performed in a serial order instead of parallel processing. Either order may be the order of processing.

  Then, the generated shot data and the calculated shift amount and shift direction information are output to the control circuit 462. Then, the drawing processing unit 450 controlled by the control circuit 462 uses the electron beam 200 and the absorber film so that the position determined to be defective is deviated from the EUV light reflection surface when performing EUV exposure. A pattern is drawn on a mask blank 101 in which another film such as a resist is further formed on the multilayer film (capping layer). Specifically, the following operation is performed. An electron beam 200 emitted from an electron gun 201 which is an example of an irradiation unit illuminates the entire first aperture 203 having a rectangular hole, for example, a rectangular hole, by an illumination lens 202. Here, the electron beam 200 is first formed into a rectangle, for example, a rectangle. Then, the electron beam 200 of the first aperture image that has passed through the first aperture 203 is projected onto the second aperture 206 by the projection lens 204. The position of the first aperture image on the second aperture 206 is deflection-controlled by the deflector 205, and the beam shape and size can be changed. As a result, the electron beam 200 is shaped. Thus, the electron beam 200 is variably shaped. In addition, the control circuit 462 corrects the position of each shot data so as to be shifted by the calculated shift amount in the calculated shift direction. As a result, the deflection amount deflected by the deflector 208 is corrected by the shift amount calculated in the calculated shift direction. Then, the electron beam 200 of the second aperture image that has passed through the second aperture 206 is focused by the objective lens 207 and deflected by the deflector 208. As a result, for example, the corrected position of the mask blank 101 on the continuously moving XY stage 405 is irradiated.

  As described above, the absorber pattern is formed on the mask blank so that the defect is hidden behind the absorber pattern. Since the obtained reflection type exposure mask has a defect hidden under the absorber pattern, there is no problem in exposing and projecting a hole pattern onto a semiconductor substrate, for example.

  Alternatively, the partial pattern data / shift amount and direction calculation unit 64 is arranged so that the absorber pattern is sufficiently separated from the phase defect when the defect PD on the mask blank is small so as not to affect the transfer by itself. The shift amount and the shift direction are calculated by shifting the pattern layout. By using such a shift amount and shift direction, in the obtained reflection type exposure mask, there is no phase defect PD in the vicinity of the absorber pattern, so that the exposure projection of the absorber pattern ABS onto the semiconductor substrate is hindered. Pattern transfer can be performed without any problems.

  As described above, according to the second embodiment, the position of the phase defect can be specified, and the positional relationship between the absorber pattern for defining the semiconductor integrated circuit and the phase defect of the mask blank can be adjusted. As a result, the frequency with which a mask blank having a phase defect can be used as a non-defective product is increased, the yield of the mask blank is greatly improved, and the cost of the reflective mask to be manufactured can be reduced.

  Here, in Embodiment 2, a configuration using an electron beam as an example of a charged particle beam has been described. However, a charged particle beam is not limited to an electron beam, and charged particles such as an ion beam may be used. I do not care.

  The embodiments have been described above with reference to specific examples. However, the present invention is not limited to these specific examples. For example, although the shift amount and the shift direction are calculated in the drawing unit 400 in the second embodiment, the present invention is not limited to this. Information on the shift amount and shift direction calculated by the user may be input. Alternatively, the drawing unit 400 may input the drawing data itself corrected by the shift amount and shift direction calculated by the user.

  In addition, although descriptions are omitted for parts and the like that are not directly required for the description of the present invention, such as a device configuration and a control method, a required device configuration and a control method can be appropriately selected and used. For example, a detailed description of the control unit configuration for controlling the mask inspection apparatus 100, the mask manufacturing apparatus 500, the inspection unit 300, and the drawing unit 400 is omitted, but the necessary control unit configuration is appropriately selected and used. Needless to say.

  In addition, all charged particle beam writing apparatuses and charged particle beam writing methods that include elements of the present invention and that can be appropriately modified by those skilled in the art are included in the scope of the present invention.

2 Device pattern area 10, 13 Light flux 11, 12, 21 Region 14, 15 Image 17, 18 Imaging point 22 Inspection stripe 23 Arrow 50 Drawing region 52, 54 ALN-mark 56, 58 Partial region 62 Defect coordinate / size acquisition unit 64 Partial pattern data / shift amount and direction calculation unit 76 Drawing data processing unit 78 Memory 100 Mask inspection device 101 Mask blank 102 XYθ table 103 Light source 104 Objective lens 105, 305 Sensor 106, 306 Sensor circuit 107, 307 Memory 108 Comparison circuit 110 Control computer 109 Storage device 111 Table control circuit 118 Monitor 119 Printer 122, 124 Lens 123 Slit plate 126 Beam splitter 128 Reflecting mirror 132, 332 Lens 172 Alignment processing unit 174 Difference calculation 176 determination unit 120 bus 200 electron beam 201 electron gun 202 illumination lens 203 first aperture 204 projection lens 205, 208 deflector 206 second aperture 207 objective lens 300 inspection unit 301 EUV mask 400 drawing unit 402 electron column 403 Drawing chamber 405 XY stage 406 Support pins 440 and 442 Storage device 410 Computer unit 450 Drawing processing unit 460 Control unit 462 Control circuit 500 Mask manufacturing device 510 Glass substrate 512 Multilayer film 514 Cap film 516 Absorber film 518 Antireflection film 540 Defect 542 , 544 areas

Claims (5)

An illumination optical system that illuminates illumination light on the surface of a mask blank on which a multilayer film that reflects extreme ultraviolet (EUV) light is formed;
First and second sensors that capture images at the same position reflected from the mask blank surface at different defocus amounts;
A determination unit that determines the presence or absence of a defect of the mask blank using the first and second optical images at the same position on the surface of the mask blank imaged with different defocus amounts;
A mask inspection apparatus comprising: For each pixel, the image processing apparatus further includes a difference calculation unit that calculates a difference between the first and second optical images.
The mask inspection apparatus according to claim 1, wherein the determination unit determines a defect for each pixel when the difference exceeds a threshold value. The first sensor captures the first optical image on the front side of the image formation point of the first optical image,
3. The mask inspection apparatus according to claim 1, wherein the second sensor images the second optical image behind the imaging point of the second optical image.   The mask inspection apparatus according to claim 1, wherein as the illumination light, deep ultraviolet (DUV) light having a wavelength of 180 to 210 nm is used. An illumination optical system that illuminates illumination light on the surface of a mask blank on which a multilayer film that reflects extreme ultraviolet (EUV) light is formed;
First and second sensors that capture images at the same position reflected from the mask blank surface at different defocus amounts;
A determination unit that determines the presence or absence of a defect of the mask blank using the first and second optical images at the same position on the surface of the mask blank imaged with different defocus amounts;
A drawing unit that draws a pattern on the mask blank on which another film is further formed using a charged particle beam so that a position determined to have a defect is deviated from the reflective surface of the EUV light when performing EUV exposure. When,
An exposure mask manufacturing apparatus comprising:

Images