JP2012058558A - Mask blank inspection method and manufacturing method of mask - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technology for detecting a minute concavo-convex defect (phase defect) of only several nm with high sensitivity in a mask blank inspection method that uses DUV light that allows usage of an inspection optical system which has high degree of completion as a mass production device and uses a lens in the atmospheric air.SOLUTION: A first image detection unit SE1 is arranged at a position where an optical path length is shorter than an imaging plane 16 of a lens L3 by XL1, consequently an expanded inspection image of an inspection light radiated area 12 on a mask blank MB is collected in a negatively defocused state. On the other hand, a second image detection unit SE2 is arranged at a position where an optical path length is longer than an imaging plane 17 of a lens L4 by XL2, consequently an expanded inspection image of an inspection light radiated area 11 on the mask blank MB is collected in a positively defocused state.

Description

  The present invention relates to a mask blank inspection method and a mask manufacturing method, and in particular, inspection of a mask blank for EUVL (Extreme Ultra Violet Lithography) using extreme ultraviolet (EUV) having a wavelength of around 13.5 nm. The present invention relates to a technique effective when applied to a method and a manufacturing method of a mask used for EUVL.

  Japanese Laid-Open Patent Publication No. 2001-174415 (Patent Document 1) and Japanese Laid-Open Patent Publication No. 2002-333313 (Patent Document 2) disclose a bright-field image as a method of irradiating a substrate with laser light and detecting foreign matter from irregularly reflected light. An inspection method using (microscopic image) is described. In particular, Patent Literature 1 and Patent Literature 2 describe a technique for discriminating whether a defect is a convex defect or a concave defect by giving asymmetry to a detection signal.

  Japanese Patent Laid-Open No. 2003-114200 (Patent Document 3) and US Application Publication No. 2004/0057107 (Patent Document 4) describe defects existing in a mask blank using inspection light having the same wavelength as EUV light used for exposure. Techniques for detecting are described. In particular, Patent Document 3 describes a technique that uses a dark field image, and Patent Document 4 describes a technique that uses a bright field image.

  Japanese Patent Laying-Open No. 2005-265736 (Patent Document 5) describes a technique of using DUV light (Deep Ultra Violet) used for inspection of a general optical mask pattern as inspection light.

JP 2001-174415 A JP 2002-333313 A JP 2003-114200 A US Application Publication No. 2004/0057107 JP 2005-265736 A

  In a semiconductor device (semiconductor device), photolithography is performed by irradiating a mask on which a circuit pattern is drawn with exposure light, and transferring the circuit pattern formed on the mask onto a semiconductor substrate (semiconductor wafer) via a reduction optical system. Formed by repeated use of technology.

  In recent years, semiconductor devices have been miniaturized, and a technique for improving the resolution by shortening the wavelength of exposure light used in the photolithography technique has been studied. For example, ArF lithography technology using an argon fluoride (ArF) excimer laser beam with a wavelength of 193 nm has been developed so far, but a lithography technology using EUV light with a wavelength of 13.5 nm which is much shorter than that. (EUVL (Extreme Ultra Violet Lithography)) is being developed.

  In a normal photolithography technique, a configuration is adopted in which exposure light is transmitted through a mask on which a circuit pattern is formed, thereby transferring the pattern formed on the mask onto a semiconductor substrate. That is, the mask is made of a transmissive material that transmits exposure light, and a light shielding pattern is formed on the transmissive material, thereby forming a desired circuit pattern on the mask. However, there is no substance transparent to EUV light in the EUV light wavelength region. That is, in a lithography technique using EUV light, a mask that transmits exposure light cannot be formed because EUV light is absorbed by a substance and does not transmit. For this reason, in the lithography technique using EUV light, an optical system that reflects the exposure light from the mask instead of transmitting the exposure light through the mask is employed.

  Specifically, in a lithography technique using EUV light, a multilayer film reflective substrate using reflection on a multilayer film composed of a molybdenum (Mo) layer and a silicon (Si) layer is used as a mask substrate (mask blank). The reflection by the multilayer film is a reflection using a kind of interference. The EUVL mask has an absorber as a low reflection region on a multilayer reflective substrate (mask blank) in which a multilayer film such as a molybdenum layer and a silicon layer is deposited on a quartz glass or a low thermal expansion glass substrate. A pattern, that is, a pattern of an EUV light absorbing film and a buffer film is formed. That is, the EUVL mask has a structure in which an absorber pattern is formed on a mask blank. When EUV light is incident on the EUVL mask configured as described above, the EUV light incident on the absorber pattern is absorbed and not reflected, while EUV light is incident on the multilayer film in the region where the absorber pattern is not formed. By doing so, EUV light is reflected from the multilayer film. Thus, by exposing the EUV light reflected from the EUVL mask onto the semiconductor substrate, the circuit pattern formed on the EUVL mask can be transferred to the resist film on the semiconductor substrate.

  Thus, it can be seen that the EUVL mask has a structure in which an absorber pattern is formed on a mask blank in which a multilayer film is formed. Here, in the EUVL mask, a minute unevenness defect (phase defect) of only a few nm formed on the mask blank also causes a transfer pattern defect. For example, an EUVL mask uses EUV light as exposure light. Since the EUV light has a wavelength of about 13.5 nm, EUV light is reflected by this uneven defect even with a minute uneven defect of only a few nm. Then, a large phase change is given to the reflected EUV light. In other words, when viewed from EUV light, a fine unevenness of only a few nanometers is not negligible, and the light intensity in an area where sufficient reflected light should be obtained by phase change and irregular reflection due to the fine unevenness. May be reduced, and the state may be the same as if the absorber pattern exists and the reflected light is absorbed. In other words, if there are minute irregularities of only a few nanometers in the reflective area of the mask blank where the absorber pattern is not formed, the fine irregularities function as if they were an absorber pattern, and a normal transfer pattern is formed. There is a risk that it will be impossible.

  On the other hand, in the ArF lithography technique using an argon fluoride (ArF) excimer laser beam having a wavelength of 193 nm, there is no problem even if there are minute uneven defects of only a few nm as described above. In other words, since the wavelength (193 nm) is sufficiently larger than the fine unevenness defect (several nm) and the transmission type mask is used instead of the reflection type mask, in a normal optical lithography technique, a fineness of only a few nm is required. Irregular defects can be ignored. As described above, the EUVL mask and the normal photolithographic mask have a large difference with respect to the defect due to the difference in the wavelength used for the exposure light and the difference between the reflective mask and the transmissive mask. I understand. That is, in the EUVL mask, a fine unevenness defect of only a few nanometers, which was not a problem in the conventional photolithographic mask, causes a pattern transfer failure. Therefore, in the EUVL mask, it is necessary to inspect whether or not a minute unevenness of only a few nm exists in the mask blank at the stage of the mask blank before forming the absorber pattern.

  For example, an inspection method using a laser beam described in Patent Document 1 or Patent Document 2 is conceivable as an inspection method for detecting a minute unevenness defect of only a few nm formed on a mask blank. However, in the inspection method using laser light, it is difficult to detect the fine unevenness defect because the fine unevenness defect of several nm is two digits or more smaller than the wavelength of the inspection light (laser light).

  On the other hand, in the inspection methods described in Patent Document 3 and Patent Document 4, since EUV light having a short wavelength is used as inspection light, it is known that both can detect a phase defect with high sensitivity. However, there is a problem that it is difficult to handle the inspection optical system. That is, the inspection optical system in the vacuum chamber that employs the EUV light source for the inspection light and the multilayer film reflection surface is difficult to handle, and there is a problem that the damage to the optical mirror cannot be said to be sufficiently small.

  On the other hand, the inspection method described in Patent Document 5 uses DUV light (wavelength of about 200 nm) as inspection light, so the degree of completion as a mass production apparatus is high, and inspection using a lens in the atmosphere Since an optical system can be used, it is an effective inspection method for a mask blank. However, since the wavelength (about 200 nm) of DUV light is about two orders of magnitude longer than the irregularities of the phase defect to be detected, the inspection method using the laser light described in Patent Document 1 and Patent Document 2 described above Similarly, there is a problem that the detection sensitivity of the phase defect is lowered.

  The object of the present invention is to provide a mask blank inspection method using DUV light that has a high degree of completion as a mass production apparatus and can use an inspection optical system using a lens in the atmosphere. The object is to provide a technique capable of detecting a defect (phase defect) with high sensitivity.

  The above and other objects and novel features of the present invention will be apparent from the description of this specification and the accompanying drawings.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  The mask blank inspection method in a typical embodiment includes (a) a step of placing a mask blank on which a multilayer film that reflects EUV light is formed, and (b) an inspection after the step (a). Irradiating the mask blank with light. (C) After the step (b), when collecting an enlarged inspection image of the inspection light irradiation area through the objective lens while scanning the inspection light irradiation area of the mask blank, Collecting the enlarged inspection image at a first predetermined amount of defocus position. Further, after the step (d) and the step (b), the objective lens is collected when the enlarged inspection image of the inspection light irradiation region is collected through the objective lens while scanning the inspection light irradiation region of the mask blank. For collecting the enlarged inspection image at a defocus position of a second predetermined amount different in sign from the first predetermined amount. Subsequently, (e) determining the presence or absence of a defect based on the enlarged inspection image collected in the step (c) and the enlarged inspection image collected in the step (d), (f) And (e) a step of displaying a determination result obtained in the step.

  Further, the mask manufacturing method according to the representative embodiment includes (a) a step of inspecting whether or not there is a defect in a mask blank on which a multilayer film that reflects EUV light is formed, and (b) the above (a). Forming the absorber pattern based on the position coordinates of the detected defect when the defect is detected in the process. At this time, the step (a) includes (a1) a step of placing the mask blank on a stage, and (a2) a step of irradiating the mask blank with inspection light after the step (a1). (A3) After the (a2) step, when collecting an enlarged inspection image of the inspection light irradiation region through the objective lens while scanning the inspection light irradiation region of the mask blank, Collecting the enlarged inspection image at a first predetermined amount of defocus position. Further, (a4) after the step (a2), when collecting the enlarged inspection image of the inspection light irradiation region through the objective lens while scanning the inspection light irradiation region of the mask blank, the objective lens The step of collecting the enlarged inspection image at a defocus position of a second predetermined amount having a sign different from that of the first predetermined amount. Subsequently, (a5) a step of determining the presence or absence of a defect based on the enlarged inspection image collected in the step (a3) and the enlarged inspection image collected in the step (a4), and (a6) A step of storing position coordinates of the defect when the defect present in the mask blank is detected as a result of performing the step (a5).

  Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows.

  In a mask blank inspection method using DUV light, which has a high degree of completion as a mass production device and can use an inspection optical system using a lens in the atmosphere, a minute unevenness defect (phase defect) of only a few nm It can be detected with high sensitivity.

It is a block diagram which shows the exposure optical system which uses EUV light. It is a figure which shows the plane structure of a mask. It is sectional drawing which shows one cross section in the device pattern area of the mask shown in FIG. It is sectional drawing which shows the state in which the phase defect is formed on the mask substrate. It is sectional drawing which shows the case where a phase defect remains in the reflective area | region between absorbers. It is a block diagram which shows the inspection optical system which takes in the defect image which exists in a mask blank in the inspection apparatus which uses DUV light as inspection light. When there is a concave phase defect on the multilayer film formed on the mask blank, it is formed on the light receiving surface of the image detector when the amount of displacement of the mask blank surface in the optical axis direction of the objective lens is changed. 6 is a graph showing the result of the light intensity distribution of each enlarged inspection image obtained by simulation analysis. When a convex phase defect exists on the multilayer film formed on the mask blank, it is formed on the light receiving surface of the image detector when the amount of shifting the mask blank surface in the optical axis direction of the objective lens is changed. 6 is a graph showing the result of the light intensity distribution of each enlarged inspection image obtained by simulation analysis. It is a figure which shows the structure of the inspection apparatus used for the inspection method of the mask blank in Embodiment 1 of this invention. In Embodiment 1, it is a figure which shows a mode that two inspection light irradiation areas are scanned over a mask blank. It is a figure which shows a mode that two inspection light irradiation area | regions are moved within the inspection area | region of one inspection stripe. 4 is a time chart for explaining a processing process with one inspection stripe in the inspection apparatus according to the first embodiment. 4 is a flowchart showing a flow of a mask blank inspection method in the first embodiment. It is a figure which shows the structure of the test | inspection apparatus in Embodiment 2. FIG. It is a figure which shows the structure of the test | inspection apparatus used for the mask blank test | inspection method in Embodiment 3. FIG. 10 is a time chart for explaining a processing process with one inspection stripe in the inspection apparatus according to the third embodiment. 10 is a flowchart showing a flow of a mask blank inspection method according to Embodiment 3. FIG. 10 is a plan view showing an example of a mask manufactured by the mask manufacturing method in the fourth embodiment. It is sectional drawing cut | disconnected by the AA line of FIG. 10 is a flowchart for explaining a mask manufacturing method according to the fourth embodiment. FIG. 10 is a plan view showing an example of a mask manufactured by the mask manufacturing method in the fourth embodiment.

  In the following embodiments, when it is necessary for the sake of convenience, the description will be divided into a plurality of sections or embodiments. However, unless otherwise specified, they are not irrelevant to each other. There are some or all of the modifications, details, supplementary explanations, and the like.

  Further, in the following embodiments, when referring to the number of elements (including the number, numerical value, quantity, range, etc.), especially when clearly indicated and when clearly limited to a specific number in principle, etc. Except, it is not limited to the specific number, and may be more or less than the specific number.

  Further, in the following embodiments, the constituent elements (including element steps and the like) are not necessarily indispensable unless otherwise specified and apparently essential in principle. Needless to say.

  Similarly, in the following embodiments, when referring to the shape, positional relationship, etc., of components, etc., unless otherwise specified, and in principle, it is considered that this is not clearly the case, it is substantially the same. Including those that are approximate or similar to the shape. The same applies to the above numerical values and ranges.

  In all the drawings for explaining the embodiments, the same members are denoted by the same reference symbols in principle, and the repeated explanation thereof is omitted. In order to make the drawings easy to understand, even a plan view may be hatched.

(Embodiment 1)
For example, ArF lithography technology using an argon fluoride (ArF) excimer laser beam having a wavelength of 193 nm has been developed. In order to cope with further pattern miniaturization, a wavelength of 13.5 nm (which is much shorter than that) Development of lithography technology (EUVL (Extreme Ultra Violet Lithography)) using EUV light of 12 nm to 15 nm) is in progress. In other words, the shorter the wavelength of the exposure light, the more finer the pattern can be transferred to the semiconductor wafer, so that the wavelength of the exposure light is shortened in order to transfer the further miniaturized pattern. It is.

  At this time, for example, in a normal photolithography technique using an argon fluoride (ArF) excimer laser beam having a wavelength of 193 nm, a pattern formed on the mask is formed by transmitting exposure light to the mask on which the circuit pattern is formed. A structure for transferring onto a semiconductor substrate is employed. That is, the mask is made of a transmissive material that transmits exposure light, and a light shielding pattern is formed on the transmissive material, thereby forming a desired circuit pattern on the mask. However, there is no substance transparent to EUV light in the EUV light wavelength region. That is, in a lithography technique using EUV light, a mask that transmits exposure light cannot be formed because EUV light is absorbed by a substance and does not transmit. For this reason, in the lithography technique using EUV light, an optical system that reflects the exposure light from the mask instead of transmitting the exposure light through the mask is employed.

  FIG. 1 is a block diagram showing an exposure optical system using EUV light. As shown in FIG. 1, the exposure optical system using EUV light includes a light source LS, an illumination optical system LOS1, a mask M, an imaging optical system LOS2, and a stage ST. The light source LS is a light source that emits EUV light (extreme ultraviolet rays) having a wavelength of 12 nm to 15 nm. The illumination optical system LOS1 is configured to receive EUV light from the light source LS, process the incident EUV light, and irradiate the mask M. Normally, the illumination optical system LOS1 is configured using a lens, but since there is no material transparent to EUV light, the illumination optical system LOS1 that uses EUV light has a reflecting mirror that reflects EUV light (for example, A concave mirror or a convex mirror). That is, the illumination optical system LOS1 is configured to process the shape of EUV light using a reflecting mirror. A pattern (for example, a circuit pattern) to be transferred to the semiconductor substrate 1S is formed on the mask M so that EUV light emitted from the illumination optical system LOS1 reflects and reflects the pattern formed on the mask M. It has become. The imaging optical system LOS2 is configured to enter EUV light reflected by the mask M and form an EUV light image on the semiconductor substrate 1S. Usually, the imaging optical system LOS2 is configured using a lens, but since there is no material transparent to EUV light, the imaging optical system LOS2 using EUV light reflects a EUV light reflecting mirror ( For example, it is composed of a concave mirror or a convex mirror. That is, the imaging optical system LOS2 is configured to collect EUV light using a reflecting mirror and form an image of the mask pattern on the semiconductor substrate 1S. The stage ST is for mounting the semiconductor substrate 1S, and is configured to arrange the semiconductor substrate 1S at a position where EUV light is imaged by the imaging optical system LOS2. The exposure optical system shown in FIG. 1 uses EUV light, and in order not to attenuate the EUV light, the exposure optical system having the light source LS, the illumination optical system LOS1, the mask M, the imaging optical system LOS2, and the stage ST is Placed in a vacuum.

  The exposure optical system in Embodiment 1 is configured as described above, and the operation thereof will be described below with reference to FIG. First, EUV light is irradiated from the light source LS. The EUV light emitted from the light source LS enters the illumination optical system LOS1. The EUV light incident on the illumination optical system LOS1 is processed by being reflected by a reflecting mirror constituting the illumination optical system LOS1, and then emitted from the illumination optical system LOS1. Subsequently, EUV light emitted from the illumination optical system LOS1 enters the mask M. The EUV light incident on the mask M is reflected by reflecting the pattern formed on the mask M. Then, the EUV light reflected by the mask M enters the imaging optical system LOS2. The EUV light incident on the imaging optical system LOS2 is processed by a reflecting mirror constituting the imaging optical system LOS2, and is formed on the semiconductor substrate 1S (specifically, formed on the semiconductor substrate 1S) on the stage ST. A pattern is formed on the resist film. That is, the circuit pattern formed on the mask M is reduced and projected onto the semiconductor substrate 1S by the imaging optical system LOS2. In this way, the circuit pattern formed on the mask M can be transferred to the resist film of the semiconductor substrate 1S.

  As described above, a pattern is formed on the mask M, and reflected light corresponding to the pattern can be emitted by making EUV light incident on this pattern. Hereinafter, the structure of the pattern formed on the mask M will be described.

  FIG. 2 is a diagram showing a planar configuration of the mask M. As shown in FIG. As shown in FIG. 2, the mask M has a rectangular shape, and alignment marks MA are formed at the four corners of the rectangular mask M. The alignment mark MA has a pattern group having a function of aligning the mask M. A device pattern area MDE is formed at the center of the mask M. In the device pattern area MDE, a circuit pattern constituting the semiconductor device is formed. Therefore, the EUV light that enters and reflects on the mask M is applied to the device pattern area MDE.

  FIG. 3 is a cross-sectional view showing a cross section in the device pattern area MDE of the mask M shown in FIG. As shown in FIG. 3, in the mask M, a multilayer film MLF is formed on a mask substrate (mask blank) MS. The mask substrate MS is made of, for example, quartz glass or a low thermal expansion material, and the multilayer film MLF has, for example, a structure in which a silicon layer and a molybdenum layer are laminated over a plurality of layers. The multilayer film MLF is a film provided to reflect EUV light incident on the mask M. That is, for example, in a multilayer film MLF composed of a silicon layer and a molybdenum layer, EUV light is reflected by utilizing multilayer film reflection that occurs when the refractive index of the silicon layer and the refractive index of the molybdenum layer are different. This multilayer film reflection is a reflection utilizing a kind of interference. Specifically, EUV light having a certain intensity or more can be reflected by interference and strengthening of light reflected from the surface of the multilayer film MLF and light reflected from the inside of the multilayer film MLF. .

  A capping layer CAP is formed on the multilayer film MLF. The capping layer CAP is a film provided to protect the multilayer film MLF, and is formed of, for example, a silicon layer or a ruthenium layer. On the capping layer CAP, the absorber ABS is formed in a pattern via the buffer layer BUF. The absorber ABS absorbs EUV light incident on the mask M so as not to be reflected. The absorber ABS is made of, for example, a tantalum material such as tantalum nitride (TaN). The absorber ABS has a thickness of, for example, 50 nm to 100 nm from the viewpoint of sufficiently absorbing EUV light. The buffer layer BUF is a film that functions as a stopper for etching performed when the absorber ABS is processed, and has a function of protecting the multilayer film MLF formed under the buffer layer BUF from etching damage. . A conductor film CF is formed on the back surface of the mask substrate MS. The conductor film CF is a film provided for electrostatic chucking the mask M on the stage.

  In the mask M configured as described above, patterning for EUV light is performed by providing a region where the multilayer film MLF is exposed and a region where the absorber ABS is formed. That is, when EUV light is incident on the mask M, the EUV light is reflected and the reflected light is emitted in the region where the multilayer film MLF is exposed. On the other hand, EUV light is absorbed and no reflected light is generated in the region where the absorber ABS is formed. For this reason, it is possible to obtain reflected light reflecting the circuit pattern by forming the circuit pattern in the region where the multilayer film MLF is exposed and the region where the absorber ABS is formed.

  Here, in the EUVL mask, a minute unevenness defect (phase defect) of only a few nm formed on a mask blank (mask substrate) also causes a transfer pattern defect. For example, FIG. 4 is a cross-sectional view showing a state where the phase defect PD is formed on the mask substrate MS. As shown in FIG. 4, when the multilayer film MLF is deposited on the mask substrate MS, if the multilayer film MLF is formed with a fine depression existing on the mask substrate MS, the mask film MS on which the multilayer film MLF is formed is formed. It can be seen that a concave phase defect PD occurs. When the mask M is manufactured by forming the buffer layer BUF and the absorber ABS while leaving the phase defect PD, for example, the phase defect PD may remain between the adjacent absorber ABS as shown in FIG. is there. FIG. 5 is a cross-sectional view showing a case where the phase defect PD remains in the reflection region between the absorbers ABS.

  As shown in FIG. 5, when the phase defect PD exists in the reflection region, the wavelength of the EUV light used for the exposure light is about 13.5 nm. Therefore, even with a minute phase defect PD of only a few nm, this phase defect When EUV light is reflected by the PD, a large phase change is given to the reflected EUV light. That is, when viewed from EUV light, a minute phase defect PD of only a few nm is not negligible, and an area in which sufficient reflected light is supposed to be obtained by phase change or irregular reflection due to the minute phase defect PD. There is a possibility that the light intensity is lowered and the state is the same as the state in which the absorber ABS is present and the reflected light is absorbed. That is, if a minute phase defect PD of only a few nm exists in the reflection region of the mask substrate MS (mask blank) where the absorber ABS is not formed, the minute phase defect PD functions as if it is an absorber ABS. There is a risk that a normal transfer pattern cannot be formed. Therefore, in the EUVL mask, whether or not a minute phase defect PD of only a few nm exists on the mask substrate MS (mask blank) at the stage of the mask substrate MS (mask blank) before forming the absorber ABS. Need to be inspected.

  First, as a method for detecting a phase defect of about several nanometers existing in the mask substrate MS, it is conceivable to use EUV light, which is also used as exposure light, as inspection light. In other words, since phase defects with a wavelength of about 13.5 nm and EUV light with a height of about several nanometers are problematic, it is possible to detect phase defects with high sensitivity by using this EUV light as inspection light. It is considered possible. However, when EUV light is used as inspection light, the following problems exist. That is, there is no substance transparent to EUV light in the EUV light wavelength region. For this reason, in the inspection using EUV light, it is necessary to use an inspection optical system in a vacuum chamber that employs an EUV light source for inspection light and a multilayer film reflecting surface. However, the inspection optical system in the vacuum chamber that employs an EUV light source for inspection light and a multilayer film reflecting surface has a problem that it is difficult to handle.

  On the other hand, it is conceivable to use DUV light (wavelength of about 180 nm to 210 nm) as inspection light. This inspection method using DUV light has a high degree of completion as a mass production apparatus, and can be used as an inspection optical system using a lens in the atmosphere. However, since the wavelength of DUV light (180 nm to 210 nm) is about two orders of magnitude longer than the irregularities of the phase defect to be detected, it is difficult to detect a phase defect having a height of about several nm with high sensitivity.

  Therefore, in the present invention, it is premised that an inspection method using DUV light is adopted, which has a high degree of completion as a mass production apparatus, can use an inspection optical system using a lens in the atmosphere, and is easy to handle. Even in the inspection method using DUV light, the device is devised to improve the detection sensitivity of phase defects having a height of several nanometers. Below, the inspection method of the mask blanks (mask substrate) which gave this device is demonstrated.

  First, knowledge that can detect a minute phase defect having a height about two orders of magnitude smaller than the wavelength of the DUV light with high sensitivity even in the inspection method using the DUV light will be described. FIG. 6 is a configuration diagram illustrating an inspection optical system that captures a defect image existing on a mask blank (mask substrate) in an inspection apparatus that uses DUV light as inspection light. As shown in FIG. 6, the inspection optical system includes a light source ILS, a beam splitter BSP, an objective lens OBL, a mask blank MB, and an image detector SE. The light source ILS is configured to emit DUV light having a wavelength of about 180 nm to 210 nm, and the DUV light BM1 emitted from the light source ILS is incident on the beam splitter BSP. Then, the DUV light BM1 incident on the beam splitter BSP is bent by the beam splitter BSP and irradiated onto a predetermined area of the mask blank MB through the objective lens OBL. At this time, a multilayer film is formed on the mask blank MB, and the surface of the mask blank MB on which the multilayer film is formed is irradiated with the DUV light BM1. That is, in FIG. 6, a multilayer film is formed on the lower surface of the mask blank MB. The DUV light BM1 irradiated to the mask blank MB is reflected, and the reflected DUV light BM2 is collected by the objective lens OBL and transmitted through the beam splitter BSP. The DUV light BM2 that has passed through the beam splitter BSP is applied to the light receiving surface of the image detector SE. At this time, the position of the image detector SE is adjusted so that an enlarged inspection image of the surface of the mask blank MB is formed on the light receiving surface of the image detector SE. That is, the surface of the mask blank MB and the light receiving surface of the image detector SE are in a conjugate positional relationship via the objective lens OBL. In this way, the enlarged inspection image on the surface of the mask blank MB can be received by the image detector SE. In this case, it is considered that the phase defect formed on the mask blank MB can be detected by analyzing the enlarged inspection image received by the image detector SE. However, in actuality, only by forming an enlarged inspection image of the surface of the mask blank MB on the light receiving surface of the image detector SE, the change in light intensity due to fine phase defects is slight, and this phase defect is detected. It is difficult.

  Therefore, the present inventor does not form an enlarged inspection image of the surface of the mask blank MB on the light receiving surface of the image detector SE, but rather determines the positional relationship between the surface of the mask blank MB and the light receiving surface of the image detector SE. The light intensity distribution of the enlarged inspection image formed on the light receiving surface of the image detector SE based on the idea that the light intensity distribution of the fine phase defect can be changed by setting the defocus state. The simulation was analyzed. The result of this simulation analysis will be described.

  FIG. 7 shows an image detector when the amount of displacement of the surface of the mask blank in the optical axis direction of the objective lens is changed when the concave phase defect PD1 exists on the multilayer film MLF formed on the mask blank. It is a graph which shows the result of having calculated | required the light intensity distribution of each expansion inspection image formed in the light-receiving surface of this by simulation analysis. Here, the depth of the concave phase defect PD1 is 2.2 nm, and the full width at half maximum is 60 nm. The numerical aperture (NA) of the objective lens is set to 0.75. Further, it is assumed that the value of the coherence factor σ indicating the degree of coherence of the illumination light (DUV light) is 1.0. When the value of the coherence factor σ is smaller than 1, the coherence of the illumination light increases and the intensity change of the image such as a defect tends to increase.

  The horizontal axis in FIG. 7 indicates the position (nm) on the mask blank, and the vertical axis in FIG. 7 indicates the light intensity of the enlarged inspection image.

  Here, in FIG. 7, the amount by which the position of the mask blank shifts from the focus position along the optical axis direction of the objective lens, that is, the defocus amount is set to −400 nm, −200 nm, 0 nm, +200 nm, and +400 nm. The light intensity distributions of the enlarged inspection images obtained at times are indicated by curve 1-1, curve 1-2, curve 1-3, curve 1-4, and curve 1-5, respectively. At this time, in this specification, when the defocus amount is +, it means a shift in the direction of increasing the optical path length, and when the defocus amount is-, it means a shift in the direction of decreasing the optical path length. .

  As shown by a curve 1-3 in FIG. 7, when the defocus amount is 0 nm, that is, when the positional relationship between the mask blank and the image detector is in the focus state, the light intensity corresponding to the phase defect PD1 is It can be seen that it is slightly darker than the light intensity of. This can be considered that when the phase defect PD1 is irradiated with DUV light, the DUV light is irregularly reflected by the concave shape of the phase defect PD1, thereby reducing the light intensity corresponding to the phase defect PD1. However, this change in light intensity is only slightly darker than the surrounding light intensity, and it is considered difficult to detect the phase defect PD1 by detecting this change in light intensity.

  On the other hand, for negative defocus amounts such as −200 nm and −400 nm, the light intensity corresponding to the phase defect PD1 is larger than the light intensity at the peripheral portion as shown by the curves 1-1 and 1-2. You can see that it is getting bigger. In this case, the change amount of the light intensity is larger than the change amount of the light intensity when the defocus amount is 0 nm. That is, when the positional relationship between the mask blank and the image detector is in a negative defocus state, the light intensity change of the phase defect PD1 becomes larger than when the positional relationship between the mask blank and the image detector is in the focused state, It can be seen that the light intensity does not change in the decreasing direction, but the light intensity increases.

  Further, for positive defocus amounts such as +200 nm and +400 nm, the light intensity corresponding to the phase defect PD1 is smaller than the light intensity at the peripheral portion as shown by the curves 1-4 and 1-5. You can see that The amount of change in light intensity in this case is larger than the amount of change in light intensity when the defocus amount is 0 nm. That is, when the positional relationship between the mask blank and the image detector is in a positive defocus state, the light intensity corresponding to the phase defect PD1 is reduced as in the case where the positional relationship between the mask blank and the image detector is in the focused state. It can be seen that the amount of change in the light intensity is larger than when the positional relationship between the mask blank and the image detector is in the focused state.

  Subsequently, FIG. 8 shows a case where the amount of shift of the surface of the mask blank in the optical axis direction of the objective lens is changed in the case where the convex phase defect PD2 exists on the multilayer film MLF formed on the mask blank. It is a graph which shows the result of having calculated | required the light intensity distribution of each expansion test | inspection image formed in the light-receiving surface of an image detector by simulation analysis. Here, the height of the convex phase defect PD2 is 2.2 nm, and the full width at half maximum is 60 nm. The numerical aperture (NA) of the objective lens is set to 0.75. Further, it is assumed that the value of the coherence factor σ indicating the degree of coherence of the illumination light (DUV light) is 1.0. When the value of the coherence factor σ is smaller than 1, the coherence of the illumination light increases and the intensity change of the image such as a defect tends to increase.

  The horizontal axis in FIG. 8 indicates the position (nm) on the mask blank, and the vertical axis in FIG. 8 indicates the light intensity of the enlarged inspection image.

  Here, in FIG. 8, the amount by which the position of the mask blank shifts from the focus position along the optical axis direction of the objective lens, that is, the defocus amount is set to +400 nm, +200 nm, 0 nm, −200 nm, and −400 nm. The light intensity distributions of the enlarged inspection images obtained at times are indicated by curves 2-1, 2-2, 2-3, 2-4, and 2-5, respectively. At this time, in this specification, when the defocus amount is + (positive), it means a shift in the direction of increasing the optical path length, and when the defocus amount is-(negative), it is in the direction of decreasing the optical path length. It shall mean a shift.

  As shown by a curve 2-3 in FIG. 8, when the defocus amount is 0 nm, that is, when the positional relationship between the mask blank and the image detector is in the focus state, the light intensity corresponding to the phase defect PD2 is It can be seen that it is slightly darker than the light intensity of. This can be considered that when the phase defect PD2 is irradiated with the DUV light, the light intensity corresponding to the phase defect PD2 is lowered by the irregular reflection of the DUV light due to the convex shape of the phase defect PD2. However, this light intensity change is only slightly darker than the surrounding light intensity, and it is considered difficult to detect the phase defect PD2 by detecting this light intensity change.

  On the other hand, for positive defocus amounts such as +200 nm and +400 nm, as indicated by curves 2-1 and 2-2, the light intensity corresponding to the phase defect PD2 is larger than the light intensity at the peripheral portion. You can see that In this case, the change amount of the light intensity is smaller than the change amount of the light intensity when the defocus amount is set to 0 nm, but does not change in the direction in which the light intensity decreases, but increases in the light intensity. It turns out that it changes.

  Further, for negative defocus amounts such as −200 nm and −400 nm, the light intensity corresponding to the phase defect PD2 is higher than the light intensity at the peripheral portion as shown by the curves 2-4 and 2-5. You can see that it is getting smaller. The amount of change in light intensity in this case is larger than the amount of change in light intensity when the defocus amount is 0 nm. That is, if the positional relationship between the mask blank and the image detector is in a negative defocus state, the light intensity corresponding to the phase defect PD2 is reduced as in the case where the positional relationship between the mask blank and the image detector is in the focused state. It can be seen that the amount of change in the light intensity is larger than when the positional relationship between the mask blank and the image detector is in the focused state.

  From the above, the following knowledge is obtained by the simulation analysis described above.

  (1) Rather than forming an enlarged inspection image of the surface of the mask blank on the light receiving surface of the image detector, the positional relationship between the surface of the mask blank and the light receiving surface of the image detector is defocused, The light intensity corresponding to the fine phase defect can be greatly changed from the light intensity corresponding to the peripheral portion.

  (2) In the positive defocus state and the negative defocus state, the light intensity change corresponding to the phase defect is in the opposite direction, so the light intensity corresponding to the phase defect in the positive defocus state and the negative defocus state are If a difference from the light intensity corresponding to the phase defect in the focus state is taken, a larger light intensity difference can be obtained.

  (3) With a concave phase defect, when the positional relationship between the surface of the mask blank and the light receiving surface of the image detector is set to a negative defocus state, the light intensity corresponding to the concave phase defect is brighter than the peripheral portion. On the other hand, when the positional relationship between the mask blank surface and the light receiving surface of the image detector is set to a positive defocus state, the light intensity corresponding to the concave phase defect is darker than the peripheral portion. Change. On the other hand, in the convex phase defect, when the positional relationship between the mask blank surface and the light receiving surface of the image detector is set to a positive defocus state, the light intensity corresponding to the convex phase defect is higher than that in the peripheral portion. On the other hand, when the positional relationship between the surface of the mask blank and the light receiving surface of the image detector is set to a negative defocus state, the light intensity corresponding to the convex phase defect becomes darker than the peripheral portion. To change. That is, the defocus dependence of light intensity corresponding to the phase defect is opposite between the concave phase defect and the convex phase defect.

  Next, a mask blank inspection method according to the first embodiment that embodies the above-described knowledge will be described with reference to the drawings. First, the configuration of an inspection apparatus that realizes the mask blank inspection method according to the first embodiment will be described. FIG. 9 is a diagram showing a configuration of an inspection apparatus used in the mask blank inspection method according to the first embodiment. As shown in FIG. 9, the light projecting system of the inspection apparatus according to the first embodiment first includes a light source ILS, a lens L1, a lens L2, an aperture AP1, a beam splitter BSP, an objective lens OBL, and a defect inspection. The mask blank MB which is the target of the above is provided.

  The light source ILS is configured to emit DUV light having a wavelength of about 180 nm to 210 nm. The DUV light BM1 emitted from the light source ILS passes through the lens L1, the aperture AP1, and the lens L2. After being bent by the beam splitter BSP, the light passes through the objective lens OBL and is irradiated onto the predetermined area ILA on the mask blank MB. Thus, the two openings formed in the aperture AP1 can be projected as images on the inspection light irradiation areas 11 and 12 in the predetermined area ILA of the mask blank MB. At this time, a multilayer film is formed on the mask blank MB, and the surface of the mask blank MB on which the multilayer film is formed is irradiated with the DUV light BM1. That is, in FIG. 9, a multilayer film is formed on the lower surface of the mask blank MB. The DUV light BM1 irradiated to the mask blank MB is reflected, and the reflected DUV light BM2 is collected by the objective lens OBL and transmitted through the beam splitter BSP. The DUV light BM2 transmitted through the beam splitter BSP is condensed on the projection area PRA. That is, the predetermined area ILA on the mask blank MB is a projectable area of the objective lens OBL, and the DUV light BM2 reflected in the predetermined area ILA is transmitted through the objective lens OBL and the beam splitter BSP, and the objective lens OBL The light is condensed on the projection area PRA which is the imaging plane. Therefore, the inspection light irradiation area 11 and the inspection light irradiation area 12 on the mask blank MB are separately projected as the enlarged inspection image 13 and the enlarged inspection image 14 in the projection area PRA. At this time, for example, the inspection light irradiation area 11 on the mask blank MB corresponds to the enlarged inspection image 13 in the projection area PRA, and the inspection light irradiation area 12 on the mask blank MB corresponds to the enlarged inspection image 14 in the projection area PRA. Correspond.

  Furthermore, the light receiving system of the inspection apparatus according to the first embodiment includes a mirror 15, a lens L3, a lens L4, a first image detector SE1, and a second image detector SE2. The light beam constituting the enlarged inspection image 14 in the projection area PRA is bent by the mirror 15, and this bent light beam is enlarged on the light receiving surface of the first image detector SE1 via the lens L3. , The mirror 15, the lens L3, and the first image detector SE1 are arranged. However, as shown in FIG. 9, the first image detector SE1 is disposed at a position where the optical path length is shorter by XL1 than the imaging plane 16 of the lens L3. At this time, the imaging surface 16 of the lens L3 is in a conjugate relationship with the enlarged inspection image 14. Therefore, the inspection light irradiation region 12, the enlarged inspection image 14, and the imaging plane 16 are in a conjugate relationship with each other.

  On the other hand, the light beam constituting the enlarged inspection image 13 in the projection area PRA forms an enlarged inspection image on the light receiving surface of the second image detector SE2 via the lens L4, and the lens L4 and the second image detector. SE2 is arranged. However, as shown in FIG. 9, the second image detector SE2 is arranged at a position where the optical path length is longer by XL2 than the imaging plane 17 of the lens L4. At this time, the imaging surface 17 of the lens L4 is in a conjugate relationship with the enlarged inspection image 13. Therefore, the inspection light irradiation region 11, the enlarged inspection image 13, and the imaging plane 17 are in a conjugate relationship with each other.

  Thus, by arranging the first image detector SE1 at a position where the optical path length is shorter by XL1 than the imaging plane 16 of the lens L3, the enlarged inspection image of the inspection light irradiation region 12 on the mask blank MB is negatively degenerated. Can be collected in focus. On the other hand, by arranging the second image detector SE2 at a position where the optical path length is longer by XL2 than the imaging plane 17 of the lens L4, the enlarged inspection image of the inspection light irradiation region 11 on the mask blank MB is positive. Can be collected in defocused state. Therefore, in the inspection apparatus according to the first embodiment, the enlarged inspection images of the adjacent inspection light irradiation region 11 and inspection light irradiation region 12 on the mask blank MB can be collected in a defocused state having different signs. it can. The first image detector SE1 and the second image detector SE2 are configured so that the arrangement position can be changed.

  Further, the first image detector SE1 and the second image detector SE2 are connected to the control circuit CNT, and are configured such that data is collected by the CPU 1 and the memory 2 constituting the control circuit CNT. Specifically, the first inspection image data is collected from the first image detector SE1, and the second inspection image data is collected from the second image detector SE2. The inspection apparatus according to the first embodiment is configured such that light intensity difference data is calculated by taking a difference between the light intensity of the collected first inspection image data and the light intensity of the second inspection image data. Has been.

  Here, in the inspection apparatus according to the first embodiment, as shown in FIG. 9, the enlarged inspection images of the inspection light irradiation region 12 on the mask blank MB are collected by the first image detector SE1, but the first The light receiving surface of the image detector SE1 is disposed at a position closer to the objective lens OBL than the imaging surface of the lens L3. That is, the positional relationship between the surface of the mask blank MB (inspection light irradiation region 12) and the first image detector SE1 is in a defocused state. Similarly, in the inspection apparatus according to the first embodiment, as shown in FIG. 9, the enlarged image of the inspection light irradiation region 11 on the mask blank MB is collected by the second image detector SE2, but the second The light receiving surface of the image detector SE2 is disposed at a position farther from the objective lens OBL than the imaging surface of the lens L4. That is, the positional relationship between the surface of the mask blank MB (inspection light irradiation region 11) and the second image detector SE2 is in a defocused state. In the first embodiment, the first inspection image data is collected from the first image detector SE1 in the defocused state, and the second inspection image data is collected from the second image detector SE2 in the defocused state. ing. Therefore, in the inspection apparatus according to the first embodiment, an enlarged inspection image on the surface of the mask blank MB is not formed on the light receiving surface of the image detector (the first image detector SE1 and the second image detector SE2). By setting the positional relationship between the surface of the mask blank MB and the light receiving surface of the image detector (first image detector SE1, second image detector SE2) to a defocused state, the light intensity corresponding to the fine phase defect It can be seen that the knowledge (1) that can be greatly changed from the light intensity corresponding to the peripheral portion is realized.

  Furthermore, in the inspection apparatus according to the first embodiment, the positional relationship between the surface of the mask blank MB (inspection light irradiation region 12) and the first image detector SE1 is in a negative defocus state, while the mask blank MB The positional relationship between the surface (inspection light irradiation region 11) and the second image detector SE2 is in a positive defocus state. Then, the first inspection image data (light intensity) from the first image detector SE1 in the negative defocus state and the second inspection from the second image detector SE2 in the positive defocus state. The light intensity difference data is calculated from the difference between the image data (light intensity). Therefore, in the positive defocus state and the negative defocus state, the light intensity change corresponding to the phase defect is in the reverse direction, so the light intensity corresponding to the phase defect in the positive defocus state and the negative The knowledge (2) that a larger light intensity difference can be obtained by taking the difference from the light intensity corresponding to the phase defect in the defocused state is also embodied in the inspection apparatus according to the first embodiment. .

  In the first embodiment, not only the above-described light intensity difference data but also the first inspection image data (light intensity) itself from the first image detector SE1 in the negative defocus state is held. In addition, the second inspection image data (light intensity) from the second image detector SE2 in the positive defocus state is also held. For this reason, according to the inspection apparatus in the first embodiment, the knowledge that the light and dark defocus dependence of the light intensity corresponding to the phase defect is reversed between the concave phase defect and the convex phase defect (3). Can also be realized.

  For example, in the case of a concave phase defect, when the positional relationship between the mask blank surface and the light receiving surface of the image detector is set to a negative defocus state, the light intensity corresponding to the concave phase defect becomes brighter than the peripheral portion. On the other hand, when the positional relationship between the mask blank surface and the light receiving surface of the image detector is set to a positive defocus state, the light intensity corresponding to the concave phase defect changes so as to be darker than the peripheral portion. To do. Therefore, the first inspection image data from the first image detector SE1 in the negative defocus state includes data that changes so that the light intensity corresponding to the specific phase defect is brighter than the peripheral portion, and If the second inspection image data from the second image detector SE2 in the positive defocus state includes data that changes so that the light intensity corresponding to the specific phase defect becomes darker than the peripheral portion, this phase defect Can be determined to be a concave phase defect.

  Similarly, in the case of a convex phase defect, when the positional relationship between the mask blank surface and the light receiving surface of the image detector is set to a positive defocus state, the light intensity corresponding to the convex phase defect is brighter than the peripheral portion. On the other hand, when the positional relationship between the mask blank surface and the light receiving surface of the image detector is set to a negative defocus state, the light intensity corresponding to the convex phase defect is darker than the peripheral portion. Change. Therefore, the first inspection image data from the first image detector SE1 in the negative defocus state includes data that changes so that the light intensity corresponding to the specific phase defect is darker than the peripheral portion, and If the second inspection image data from the second image detector SE2 in the positive defocus state includes data that changes so that the light intensity corresponding to the specific phase defect becomes brighter than the peripheral portion, this phase defect Can be determined as a convex phase defect.

  From the above, the configuration of the inspection apparatus according to the first embodiment is a configuration necessary for realizing the above-described findings (1) to (3). According to this, it can be seen that the above-described findings (1) to (3) can actually be realized.

  Here, in the first embodiment, since the arrangement positions of the first image detector SE1 and the second image detector SE2 are moved, the surface of the mask blank MB (inspection light irradiation region 12) and The positional relationship between the first image detector SE1 can be in a negative defocus state, and the positional relationship between the surface of the mask blank MB (inspection light irradiation region 11) and the second image detector SE2 is in a positive defocus state. It can be.

  For example, in the simulation analysis described above, the defocus state is realized by shifting the position of the mask blank instead of the image detector. However, when the defocus state is formed by shifting the position of the mask blank, one The inspection light irradiation region 11 and the inspection light irradiation region 12 formed on the mask blank are shifted in the same direction. That is, in the method of forming the defocused state by fixing the position of the image detector on the imaging plane and shifting the position of the mask blank, the surface of the mask blank MB (inspection light irradiation region 12) and the first image detector. If the positional relationship of SE1 is in a negative defocus state, the positional relationship between the surface of the mask blank MB (inspection light irradiation region 11) and the second image detector SE2 is also in a negative defocus state. That is, in the method of forming the defocused state by fixing the position of the image detector on the imaging plane and shifting the position of the mask blank, the surface of the mask blank MB (inspection light irradiation region 12) and the first image detector. The positional relationship between SE1 can be set to a negative defocus state, and the positional relationship between the surface of the mask blank MB (inspection light irradiation region 11) and the second image detector SE2 can be set to a positive defocus state. It is not possible.

  In contrast, when the mask blank MB is fixed and the arrangement positions of the first image detector SE1 and the second image detector SE2 are moved in different directions as in the first embodiment, the mask blank is used. The positional relationship between the surface of the MB (inspection light irradiation region 12) and the first image detector SE1 can be in a negative defocus state, and the surface of the mask blank MB (inspection light irradiation region 11) and the second image detection. The positional relationship of the device SE2 can be set to a positive defocus state.

Further, when the mask blank MB is fixed and the arrangement positions of the first image detector SE1 and the second image detector SE2 are moved in different directions as in the first embodiment, there are further advantages. is there. For example, as described in the simulation analysis, the positional relationship between the mask blank MB and the image detector (the first image detector SE1 and the second image detector SE2) is set to a defocus state in which the light intensity of the phase defect is increased. For this reason, when the mask blank is moved, it is necessary to move the mask blank with an accuracy of several hundred nm, which is difficult to realize. On the other hand, when moving the first image detector SE1 and the second image detector SE2, for example, the movement amount of the first image detector SE1 is XL1, the movement amount of the second image detector SE2 is XL2, and the objective lens OBL And the magnification product of the lens L3 (several hundred times) and the product of the magnification of the objective lens OBL and the lens L4 (several hundred times), XL1 / (magnification) ² and XL2 / (magnification) ² are on the mask blank MB. Defocus amount. Therefore, even when the defocus amount of the mask blank MB is set to several hundred nm, XL1 that is the movement amount of the first image detector SE1 and XL2 that is the movement amount of the second image detector SE2 are on the order of several tens of mm. Become. Therefore, as in the first embodiment, the configuration in which the first image detector SE1 and the second image detector SE2 are moved has an advantage that a defocus state that increases the light intensity of the phase defect can be easily realized. It is.

  The inspection apparatus according to the first embodiment is configured as described above. A mask blank inspection method using this inspection apparatus will be described below with reference to the drawings. The mask blank inspection method according to the first embodiment will be clarified by describing the operation of the above-described inspection apparatus.

  First, a method for scanning the inspection light irradiation region 11 and the inspection light irradiation region 12 over the entire surface of the mask blank MB will be described. FIG. 10 is a diagram showing how the inspection light irradiation region 11 and the inspection light irradiation region 12 are scanned over the mask blank MB in the first embodiment. As shown in FIG. 10, the inspection area of the mask blank MB is divided into strip-like inspection stripes SP1 having a width of about 200 μm. The stage on which the mask blank MB is mounted is scan-controlled so that the inspection light irradiation region 11 and the inspection light irradiation region 12 move within one inspection stripe SP1. Specifically, in FIG. 10, the stage on which the mask blank MB is mounted is controlled so that the inspection light irradiation region 11 and the inspection light irradiation region 12 move from the left side to the right side in one inspection stripe SP1. Accordingly, the inspection light irradiation region 11 and the inspection light irradiation region 12 can be scanned within the region of one inspection stripe SP1. When the scanning of the inspection light irradiation region 11 and the inspection light irradiation region 12 is completed within one inspection stripe SP1, the inspection light irradiation region 11 and the inspection light irradiation region are sequentially sequentially performed within the region of the next inspection stripe. 12 is scanned. In this way, the inspection light irradiation region 11 and the inspection light irradiation region 12 can be scanned over the entire inspection region of the mask blank MB.

  Here, as shown in FIG. 9, the DUV light irradiated to the inspection light irradiation region 11 in the region of one inspection stripe SP1 passes through the objective lens OBL and the lens L4 and is received by the second image detector SE2. Incident on the surface. That is, the enlarged inspection image IMA1 corresponding to the inspection light irradiation region 11 is projected in a positive defocused state on the light receiving surface of the second image detector SE2. Similarly, the DUV light irradiated on the inspection light irradiation region 12 in the region of one inspection stripe SP1 passes through the objective lens OBL and the lens L3 and enters the light receiving surface of the first image detector SE1. That is, the enlarged inspection image IMA2 corresponding to the inspection light irradiation region 12 is projected in a negative defocus state on the light receiving surface of the first image detector SE1.

  At this time, as shown in FIG. 10, the enlarged inspection image IMA1 is taken into the second image detector SE2 as the light intensity in the plurality of pixels PIX1. That is, the second image detector SE2 includes a plurality of pixels PIX1, and is configured to capture the light intensity of the enlarged inspection image IMA1 in units of the plurality of pixels PIX1. Similarly, the enlarged inspection image IMA2 is captured by the first image detector SE1 as the light intensity in the plurality of pixels PIX2. That is, the first image detector SE1 includes a plurality of pixels PIX2, and is configured to capture the light intensity of the enlarged inspection image IMA2 in units of the plurality of pixels PIX2. The first image detector SE1 and the second image detector SE2 are composed of, for example, a TDI (Time Delay Integration) sensor. This TDI sensor is a sensor configured to transfer charges in synchronization with scanning of a stage on which a mask blank MB is mounted. As described above, the inspection light irradiation region 11 on the mask blank MB is projected as the enlarged inspection image IMA1 on the light receiving surface of the second image detector SE2 in a positive defocus state, and the projected enlarged inspection image IMA1 is It is taken in by the TDI sensor constituting the second image detector SE2 and stored as second inspection image data. Similarly, the inspection light irradiation area 12 on the mask blank MB is projected as a magnified inspection image IMA2 on the light receiving surface of the first image detector SE1 in a negative defocus state, and the projected magnified inspection image IMA2 is the first image. It is taken in by a TDI sensor constituting the detector SE1 and stored as first inspection image data.

  FIG. 11 is a diagram showing how the inspection light irradiation area 11 and the inspection light irradiation area 12 are moved within the inspection area of one inspection stripe SP1. As shown in FIG. 11, the state indicated by T <b> 1 indicates a state immediately before the stage on which the mask blank MB is mounted starts scanning and the left end of the inspection stripe SP <b> 1 approaches the inspection light irradiation region 12. The state indicated by T2 indicates a state immediately before the stage scan advances and the left end of the inspection stripe SP1 approaches the inspection light irradiation region 11. Further, the state indicated by T3 is a state in which scanning of the stage is completed, and the inspection light irradiation region 11 has exceeded the right end of the inspection stripe SP1. In this way, by scanning the stage on which the mask blank MB is mounted, the inspection light irradiation region 11 and the inspection light irradiation region 12 can be moved over the inspection region of one inspection stripe SP1.

  Next, a processing process using one inspection stripe will be described with reference to a time chart shown in FIG. FIG. 12 is a time chart for explaining the processing steps in one inspection stripe in the inspection apparatus according to the first embodiment. 9, FIG. 10 and FIG. 12, first, scanning of the mask blank MB is started at time t1, and the first image detector SE1 and the second image detector SE2 are set in a standby state. When the inspection light irradiation region 12 reaches the left end of the inspection stripe SP1 at time t2, the positional relationship between the position of the stage and the enlarged inspection image IMA2 captured by the first image detector SE1 is stored, and the control circuit CNT The enlarged inspection image IMA2 projected on the light receiving surface of the first image detector SE1 is captured as the first inspection image data based on the control of the CPU 1 in the first image detector SE1, and the captured first inspection image data is stored in the first memory (memory 2). (Some).

  Next, when the inspection light irradiation region 11 reaches the left end of the inspection stripe SP1 at time t3, the positional relationship between the position of the stage and the enlarged inspection image IMA1 captured by the second image detector SE2 is stored, and the control circuit Based on the control of the CPU 1 in the CNT, the enlarged inspection image IMA1 projected on the light receiving surface of the second image detector SE2 is captured as second inspection image data, and the captured second inspection image data is stored in the second memory (memory 2). Part of).

  Then, after the first inspection image data and the second inspection image data are continuously captured over the entire area of the inspection stripe SP1, the first inspection image data is captured by the first image detector SE1 at time t4. At the time t5, the capture of the second inspection image data by the second image detector SE2 is completed. As described above, the first inspection image data and the second inspection image data are captured in one stripe (one inspection stripe SP1) from time t2 to time t5. Then, at time t6, the scanning of the mask blank MB is completed to prepare for the transition to the next inspection stripe.

  Here, between the time t3 and the time t5, the second inspection image data captured by the second image detector SE2 is taken into the second memory, and the first image detector SE1 previously stored from the first memory is used. The captured first inspection image data is sequentially read out. Then, light intensity difference data (absolute value) is calculated by calculating a difference in light intensity between the first inspection image data and the second inspection image data at the same position (same location) of the mask blank MB.

  Since the first inspection image data and the second inspection image data are data collected in a defocused state with different signs, if there is a phase defect, for example, as shown in FIGS. As for the light intensity corresponding to the defect, one is darker than the peripheral portion and the other is brighter than the peripheral portion. Therefore, the light intensity difference data is calculated by calculating the difference in light intensity between the first inspection image data and the second inspection image data, and the absolute value of the calculated light intensity difference data is a predetermined threshold set in advance. When the value is exceeded, it is determined that a phase defect exists. In this way, it is possible to detect a phase defect present in one inspection stripe SP1.

  The mask blank inspection method according to the first embodiment is as described above, and the mask blank inspection method according to the first embodiment will be described below again using a flowchart. FIG. 13 is a flowchart showing the flow of the mask blank inspection method according to the first embodiment. First, as shown in FIG. 13, in the inspection area of the mask blank MB, the inspection stripe SP1 to be inspected first is designated (S101). Thereafter, scanning of the stage on which the mask blank MB is mounted is started (S102).

  Next, when the inspection light irradiation area 12 irradiated with DUV light enters the inspection stripe SP1, the first inspection image data is captured by the first image detector SE1, and the captured first inspection image data is stored in the first inspection image data. Store in one memory (S103). Subsequently, when the inspection light irradiation region 11 irradiated with the DUV light enters the inspection stripe SP1 by continuous scanning of the mask blank MB, the second inspection image data is captured by the second image detector SE2. The captured second inspection image data is stored in the second memory. Further, the first inspection image data collected by the first image detector SE1 is sequentially read from the first memory (S104).

  Thereafter, the first inspection image data and the second inspection image data in the same area of the mask blank MB are compared. This comparison is performed for each pixel corresponding to the same position. Then, light intensity difference data (absolute value) is calculated by calculating a difference in light intensity between the first inspection image data and the second inspection image data in the same region (S105).

  Subsequently, the calculated absolute value of the light intensity difference data is compared with a preset threshold value (S106). If the light intensity difference data is smaller than the threshold value, it is determined that no phase defect exists. Then, the process proceeds to step S108. On the other hand, if there is a pixel whose absolute value of the light intensity difference data exceeds the threshold value, it is determined that a phase defect exists, and defect information (defect size, defect position, defect shape, etc.) is displayed and stored. This is performed (S107). For example, the coordinates where the phase defect exists can be appropriately determined based on the position of a predetermined absorber pattern in a mark provided on the mask blank MB or a mask on which an absorber is formed. When the absolute value of the light intensity difference data exceeds a threshold value over a plurality of consecutive pixels, the defect size and the defect contour can be grasped.

  Further, in the first embodiment, since the first inspection image data itself and the second inspection image data itself are also stored, from the magnitude relationship of the light intensity between the first inspection image data and the second inspection image data. It can also be determined whether the phase defect has a concave shape or a phase defect having a convex shape. That is, the defocus dependence of light intensity corresponding to the phase defect is reversed between the concave phase defect and the convex phase defect, so the first inspection image data itself and the second inspection image data itself are The defect shape can also be grasped by analysis.

  Then, it is determined whether all the processes on one inspection stripe SP1 (one stripe) have been completed (S108). If not completed, the process returns to S103 to continue the inspection in one inspection stripe SP1. On the other hand, when all the processes in one inspection stripe SP1 are completed, it is determined whether the defect inspection of all stripes is completed (S109). If the defect inspection for all stripes has not been completed, a new inspection stripe is designated (S110), the mask blank MB is placed at the designated stripe start position, and the defect inspection is repeated. On the other hand, when the defect inspection of all stripes is completed, the defect inspection of the mask blank MB is completed. In this manner, the mask blank inspection method according to the first embodiment can be performed.

  From the above, according to the mask blank inspection method in the first embodiment, DUV light having a high degree of completion as a mass production apparatus and capable of using an inspection optical system using a lens in the atmosphere is used. Even in the mask blank inspection method, fine irregularities (phase defects) of only a few nm can be detected with high sensitivity. In other words, according to the first embodiment, DUV light can be used, so that the completion as a mass production apparatus is high, and an inspection optical system using lenses in the atmosphere can be used. As a result, it is possible to inspect a phase defect present in a mask blank for EUVL using an existing inspection apparatus that is easy to handle. And in this Embodiment 1, since the knowledge (1)-(3) mentioned above is embodied, even if it uses DUV light, even if it is a phase defect which has an unevenness | corrugation smaller by about 2 digits than the wavelength of DUV light The remarkable effect that it can detect with high sensitivity can be acquired.

  In the first embodiment, as shown in FIG. 9, the defocus state is realized with the mask blank MB by adjusting the positions of the first image detector SE1 and the second image detector SE2. Therefore, the following advantages can also be obtained. That is, the first embodiment is characterized in that the phase defect detection sensitivity is improved by collecting the image data of the phase defect in the defocused state, but the detection sensitivity is improved depending on the size of the phase defect. The defocus amount is considered to be different. Accordingly, in the configuration in which the defocus state is realized by adjusting the positions of the first image detector SE1 and the second image detector SE2, the first image detector SE1 and the second image are selected according to the size of the target phase defect. Since the defocus amount at the detector SE2 can be easily adjusted, there is an advantage that the target phase defect can be inspected in an optimum state.

  Furthermore, when the value of the coherence factor σ indicating the degree of coherence of the illumination light is smaller than 1, the coherence of the illumination light increases and the intensity change of the image such as a phase defect tends to increase. Therefore, also in the first embodiment, the detection sensitivity of the phase defect can be further improved by devising the diaphragm and using an optical system that makes the coherence factor σ smaller than 1.

  In the first embodiment, the method for inspecting the mask blank MB has been described. However, the technical idea of the first embodiment is not limited to this, and a mask in which the absorber ABS is partially formed on the mask blank MB. M can also be applied to defect inspection in the region of the mask M where the absorber ABS is removed and the multilayer film reflecting surface is exposed.

(Embodiment 2)
In the first embodiment, the example in which the defocus state is realized by adjusting the arrangement positions of the first image detector SE1 and the second image detector SE2 has been described. In the second embodiment, the first embodiment An example in which a defocus state is realized by inserting a lens in front of the image detector SE1 and the second image detector SE2 will be described.

  FIG. 14 is a diagram illustrating a configuration of the inspection apparatus according to the second embodiment. The configuration of the inspection apparatus in the second embodiment shown in FIG. 14 is substantially the same as the configuration of the inspection apparatus in the first embodiment shown in FIG.

  As shown in FIG. 14, in the second embodiment, a concave lens CAL is inserted between the lens L3 and the first image detector SE1. That is, the first image detector SE1 is arranged so that the light receiving surface comes to the imaging surface 16 of the lens L3, but a concave lens CAL is inserted between the lens L3 and the first image detector SE1. . For this reason, the imaging surface 16 of the lens L3 is shifted to a position farther from the objective lens OBL and the lens L3 than the light receiving surface of the first image detector SE1 by the concave lens CAL. Therefore, the first image detector SE1 collects an enlarged inspection image in a defocused state in which the imaging surface 16 of the enlarged inspection image 14 is formed at a position farther from the objective lens OBL and the lens L3 than the light receiving position. Become.

  On the other hand, in the second embodiment, a convex lens CVL is inserted between the lens L4 and the second image detector SE2. In other words, the second image detector SE2 is arranged such that the light receiving surface comes to the imaging surface 17 of the lens L4, but the convex lens CVL is inserted between the lens L4 and the second image detector SE2. . For this reason, the imaging surface 17 of the lens L4 is shifted to a position closer to the objective lens OBL and the lens L4 than the light receiving surface of the second image detector SE2 by the convex lens CVL. Therefore, the second image detector SE2 collects an enlarged inspection image in a defocused state in which the imaging surface 17 of the enlarged inspection image 13 is formed at a position closer to the objective lens OBL and the lens L4 than the light receiving position. Become.

  From the above, even in the mask blank inspection method according to the second embodiment, the mask blank using DUV light that has a high degree of completion as a mass production apparatus and can use an inspection optical system using a lens in the atmosphere. In this inspection method, it is possible to detect a minute unevenness defect (phase defect) of only a few nm with high sensitivity. That is, since the DUV light can also be used in the second embodiment, the inspection optical system using a lens in the atmosphere can be used with a high degree of completion as a mass production apparatus. As a result, it is possible to inspect a phase defect present in a mask blank for EUVL using an existing inspection apparatus that is easy to handle. And in this Embodiment 2, since the knowledge (1)-(3) mentioned above is embodied, even if it uses DUV light, even if it is a phase defect which has an unevenness | corrugation smaller by about 2 digits than the wavelength of DUV light The remarkable effect that it can detect with high sensitivity can be acquired.

  In particular, in the second embodiment, defocusing can be easily performed only by inserting the concave lens CAL and the convex lens CVL without adjusting the arrangement position of the first image detector SE1 or the arrangement position of the second image detector SE2. The advantage that the state can be realized is obtained.

(Embodiment 3)
In the first embodiment, the inspection method using two inspection light irradiation regions (the inspection light irradiation region 11 and the inspection light irradiation region 12) has been described. However, in the third embodiment, only one inspection light region is used. The inspection method to be used will be described.

  FIG. 15 is a diagram showing a configuration of an inspection apparatus used in the mask blank inspection method according to the third embodiment. As shown in FIG. 15, the light projection system of the inspection apparatus according to the third embodiment first has a light source ILS, a lens L1, a lens L2, an aperture AP2, a beam splitter BSP, an objective lens OBL, and a defect inspection. The mask blank MB which is the target of the above is provided.

  The light source ILS is configured to emit DUV light having a wavelength of about 180 nm to 210 nm, and the DUV light BM1 emitted from the light source ILS passes through the lens L1, the aperture AP2, and the lens L2. After being bent by the beam splitter BSP, the light passes through the objective lens OBL and is irradiated onto the predetermined area ILA on the mask blank MB. Here, in the third embodiment, one opening formed in the aperture AP2 can be projected as an image on the inspection light irradiation region 20 in the predetermined region ILA of the mask blank MB. At this time, a multilayer film is formed on the mask blank MB, and the surface of the mask blank MB on which the multilayer film is formed is irradiated with the DUV light BM1. That is, in FIG. 15, a multilayer film is formed on the lower surface of the mask blank MB. The DUV light BM1 irradiated to the mask blank MB is reflected, and the reflected DUV light BM2 is collected by the objective lens OBL and transmitted through the beam splitter BSP. The DUV light BM2 transmitted through the beam splitter BSP is condensed on the projection area PRA. That is, the predetermined area ILA on the mask blank MB is a projectable area of the objective lens OBL, and the DUV light BM2 reflected in the predetermined area ILA is transmitted through the objective lens OBL and the beam splitter BSP, and the objective lens OBL The light is condensed on the projection area PRA which is the imaging plane. Therefore, the inspection light irradiation area 20 on the mask blank MB is projected as an enlarged inspection image 21 in the projection area PRA.

  Further, the light receiving system of the inspection apparatus according to the third embodiment includes a convex lens CVL1, a half mirror HFM, a convex lens CVL2, a convex lens CVL3, a first image detector SE1, and a second image detector SE2. The light beam constituting the enlarged inspection image 21 in the projection area PRA described above is converted into parallel light by the convex lens CVL1, and part of the light beam converted into parallel light is reflected by the half mirror HFM. The light beam reflected by the half mirror HFM is condensed on the image plane 22 by the convex lens CVL2. That is, the enlarged inspection image 21 is formed on the imaging surface 22 by the convex lens CVL1, the half mirror HFM, and the convex lens CVL2. Here, as shown in FIG. 15, the first image detector SE1 is arranged at a position where the optical path length is shorter by XL1 than the imaging surface 22 of the convex lens CVL2. At this time, the imaging surface 22 of the convex lens CVL2 is in a conjugate relationship with the enlarged inspection image 21. Therefore, the inspection light irradiation region 20, the enlarged inspection image 21, and the imaging plane 22 are in a conjugate relationship with each other.

  On the other hand, the light beam constituting the enlarged inspection image 21 in the projection area PRA is converted into parallel light by the convex lens CVL1, and a part of the light beam converted into the parallel light is transmitted through the half mirror HFM. The light beam transmitted through the half mirror HFM is condensed on the image plane 23 by the convex lens CVL3. That is, the enlarged inspection image 21 is imaged on the imaging plane 23 by the convex lens CVL1, the half mirror HFM, and the convex lens CVL3. Here, as shown in FIG. 15, the second image detector SE2 is disposed at a position where the optical path length is longer by XL2 than the imaging plane 23 of the convex lens CVL3. At this time, the imaging surface 23 of the convex lens CVL3 is in a conjugate relationship with the enlarged inspection image 21. Therefore, the inspection light irradiation region 20, the enlarged inspection image 21, and the imaging plane 23 are in a conjugate relationship with each other.

  Thus, by disposing the first image detector SE1 at a position where the optical path length is shorter by XL1 than the imaging surface 22 of the convex lens CVL2, the enlarged inspection image 21 of the inspection light irradiation region 20 on the mask blank MB is negatively displayed. Can be collected in defocused state. On the other hand, the second image detector SE2 is arranged at a position where the optical path length is longer by XL2 than the imaging plane 23 of the convex lens CVL3, thereby correcting the enlarged inspection image 21 of the inspection light irradiation region 20 on the mask blank MB. Can be collected in a defocused state. Therefore, in the inspection apparatus according to the third embodiment, the enlarged inspection images of the inspection light irradiation region 20 on the mask blank MB can be collected in a defocused state with different signs. The first image detector SE1 and the second image detector SE2 are configured so that the arrangement position can be changed.

  Further, the first image detector SE1 and the second image detector SE2 are connected to the control circuit CNT, and are configured such that data is collected by the CPU 1 and the memory 2 constituting the control circuit CNT. Specifically, the first inspection image data is collected from the first image detector SE1, and the second inspection image data is collected from the second image detector SE2. The inspection apparatus according to the first embodiment is configured such that light intensity difference data is calculated by taking a difference between the light intensity of the collected first inspection image data and the light intensity of the second inspection image data. Has been.

  Here, the feature of the third embodiment is that the enlarged inspection image 21 is collected with respect to one inspection light irradiation region 20 in both a negative defocus state and a positive defocus state. For example, in the first embodiment, two inspection light irradiation regions 11 and an inspection light irradiation region 12 are provided, and an enlarged inspection image corresponding to the inspection light irradiation region 11 is collected in a negative defocus state (first inspection). Image data), an enlarged inspection image corresponding to the inspection light irradiation region 12 is collected in a positive defocus state (second inspection image data). Therefore, in order to calculate the light intensity difference at the same position of the mask blank MB, the first inspection image data and the second inspection image data at the same time (the positions at which the first inspection image data and the second inspection image data are different). In this case, it is necessary to compare the first inspection image data and the second inspection image data having a time difference, which complicates the data processing process. On the other hand, in the third embodiment, the enlarged inspection image 21 can be collected with respect to one inspection light irradiation region 20 in both a negative defocus state and a positive defocus state. That is, in the third embodiment, the first inspection image data and the second inspection image data are collected simultaneously for one inspection light irradiation region 20. Therefore, in the third embodiment, in order to calculate the light intensity difference at the same position of the mask blank MB, the first inspection image data and the second inspection image data at the same time may be compared, and the data processing step The advantage that can be simplified is obtained.

  The inspection apparatus according to the third embodiment is configured as described above, and a mask blank inspection method using this inspection apparatus will be described below with reference to the drawings. The mask blank inspection method according to the third embodiment will be clarified by describing the operation of the above-described inspection apparatus.

  First, the processing steps in one inspection stripe will be described using the time chart shown in FIG. FIG. 16 is a time chart for explaining the processing steps in one inspection stripe in the inspection apparatus according to the third embodiment. 15 and 16, scanning of the mask blank MB is started at time t1, and the first image detector SE1 and the second image detector SE2 are set in a standby state. At time t2, when the inspection light irradiation region 20 reaches the left end of the inspection stripe, an enlarged inspection image projected on the light receiving surface of the first image detector SE1 is controlled based on the control of the CPU 1 in the control circuit CNT. The first inspection image data is captured as the first inspection image data, and the captured first inspection image data is stored in the first memory (a part of the memory 2). At the same time, the enlarged inspection image projected on the light receiving surface of the second image detector SE2 is captured as second inspection image data, and the captured second inspection image data is stored in the second memory (a part of the memory 2).

  Then, after the first inspection image data and the second inspection image data are simultaneously and continuously captured over the entire area of the inspection stripe, the first inspection image data is captured by the first image detector SE1 at time t3. Then, the capture of the second inspection image data by the second image detector SE2 is finished. As described above, the first inspection image data and the second inspection image data are captured in one stripe (one inspection stripe) from time t2 to time t3. Then, at time t4, the scanning of the mask blank MB is completed to prepare for the transition to the next inspection stripe.

  Here, between time t2 and time t3, by calculating the difference in light intensity between the first inspection image data and the second inspection image data at the same position (same location) of the mask blank MB, light intensity difference data ( (Absolute value) is calculated. Specifically, in the third embodiment, by calculating the difference in light intensity between the first inspection image data and the second inspection image data captured at the same time, light intensity difference data at the same position of the mask blank MB ( (Absolute value) can be calculated.

  Since the first inspection image data and the second inspection image data are data collected in a defocused state with different signs, if there is a phase defect, for example, as shown in FIGS. As for the light intensity corresponding to the defect, one is darker than the peripheral portion and the other is brighter than the peripheral portion. Therefore, the light intensity difference data is calculated by calculating the difference in light intensity between the first inspection image data and the second inspection image data, and the absolute value of the calculated light intensity difference data is a predetermined threshold set in advance. When the value is exceeded, it is determined that a phase defect exists. In this way, phase defects existing in one inspection stripe can be detected.

  The mask blank inspection method according to the third embodiment is as described above, and the mask blank inspection method according to the third embodiment will be described below again using a flowchart. FIG. 17 is a flowchart showing the flow of the mask blank inspection method according to the third embodiment. First, as shown in FIG. 17, an inspection stripe to be inspected first is designated in the inspection area of the mask blank MB (S201). Thereafter, scanning of the stage on which the mask blank MB is mounted is started (S202).

  Next, when the inspection light irradiation area 20 irradiated with DUV light enters the inspection stripe, the first image detector SE1 captures the first inspection image data, and the captured first inspection image data is the first. Store in memory. At the same time, the second inspection image data is captured by the second image detector SE2, and the captured second inspection image data is stored in the second memory (S203).

  Thereafter, the first inspection image data and the second inspection image data at the same time of the mask blank MB are compared. This comparison is performed for each pixel corresponding to the same position. Then, the light intensity difference data (absolute value) at the same position of the mask blank MB can be calculated by calculating the difference in light intensity between the first inspection image data and the second inspection image data at the same time (S204). ).

  Subsequently, the calculated absolute value of the light intensity difference data is compared with a preset threshold value (S205). If the light intensity difference data is smaller than the threshold value, it is determined that there is no phase defect. Then, the process proceeds to step S207. On the other hand, if there is a pixel whose absolute value of the light intensity difference data exceeds the threshold value, it is determined that a phase defect exists, and defect information (defect size, defect position, defect shape, etc.) is displayed and stored. This is performed (S206). For example, the coordinates where the phase defect exists can be appropriately determined based on the position of a predetermined absorber pattern in a mark provided on the mask blank MB or a mask on which an absorber is formed. When the absolute value of the light intensity difference data exceeds a threshold value over a plurality of consecutive pixels, the defect size and the defect contour can be grasped.

  Further, in the third embodiment, since the first inspection image data itself and the second inspection image data itself are also stored, from the magnitude relationship of the light intensity between the first inspection image data and the second inspection image data. It can also be determined whether the phase defect has a concave shape or a phase defect having a convex shape. That is, the defocus dependence of light intensity corresponding to the phase defect is reversed between the concave phase defect and the convex phase defect, so the first inspection image data itself and the second inspection image data itself are The defect shape can also be grasped by analysis.

  Then, it is determined whether or not all the processes on one inspection stripe (one stripe) have been completed (S207), and if not completed, the process returns to S203 to continue the inspection within one inspection stripe. On the other hand, when all the processes for one inspection stripe are completed, it is determined whether the defect inspection for all stripes is completed (S208). If defect inspection has not been completed for all stripes, a new inspection stripe is designated (S209), a mask blank MB is placed at the designated stripe start position, and defect inspection is repeated. On the other hand, when the defect inspection of all stripes is completed, the defect inspection of the mask blank MB is completed. In this way, the mask blank inspection method according to the third embodiment can be carried out.

  From the above, according to the mask blank inspection method of the third embodiment, DUV light having a high degree of completion as a mass production apparatus and capable of using an inspection optical system using a lens in the atmosphere is used. Even in the mask blank inspection method, fine irregularities (phase defects) of only a few nm can be detected with high sensitivity. That is, according to the third embodiment, since DUV light can be used, the inspection optical system using a lens in the atmosphere can be used with a high degree of completion as a mass production apparatus. As a result, it is possible to inspect a phase defect present in a mask blank for EUVL using an existing inspection apparatus that is easy to handle. And in this Embodiment 3, since the knowledge (1)-(3) mentioned above is embodied, even if it uses DUV light, even if it is a phase defect which has an unevenness | corrugation smaller by about 2 digits than the wavelength of DUV light The remarkable effect that it can detect with high sensitivity can be acquired.

  In particular, in the third embodiment, the first inspection image data and the second inspection image data are collected simultaneously for one inspection light irradiation region 20. Therefore, in the third embodiment, in order to calculate the light intensity difference at the same position of the mask blank MB, the first inspection image data and the second inspection image data at the same time may be compared, and the data processing step The advantage that can be simplified is obtained.

(Embodiment 4)
In the fourth embodiment, a mask manufacturing method for manufacturing a mask for EUVL using the mask blank inspected by the inspection method described in the first to third embodiments will be described with reference to the drawings.

  FIG. 18 is a plan view showing an example of a mask manufactured by the mask manufacturing method according to the fourth embodiment. In FIG. 18, an absorber ABS is formed on the surface of the mask M, and a part of the absorber ABS is removed to form a hole pattern HP. That is, this mask M is a mask for forming a hole pattern on a semiconductor substrate (semiconductor wafer). The feature of this mask M is to inspect whether there is a phase defect in the mask blank state before forming the mask M, and even if a phase defect exists on this mask blank, as shown in FIG. The mask M is formed so that the phase defect PD is covered with the absorber ABS. Thereby, according to the mask M in the fourth embodiment, since the phase defect PD is covered with the absorber ABS, for example, the hole pattern HP is exposed and projected onto the semiconductor substrate (semiconductor wafer). However, it is possible to prevent transfer failure due to the presence of the phase defect PD.

  Specifically, FIG. 19 is a cross-sectional view taken along line AA in FIG. As shown in FIG. 19, in the mask M in the fourth embodiment, a conductor film CF is formed on the back surface (lower surface) of the mask substrate (mask blank) MS, and the main surface of the mask substrate (mask blank) MS ( A multilayer film MLF is formed on the upper surface. A capping layer CAP is formed on the multilayer film MLF, and a buffer layer BUF is formed on the capping layer CAP. Absorber ABS is formed on this buffer layer BUF. At this time, the phase defect PD is formed on the main surface (front surface) of the mask substrate (mask blank) MS, and it can be seen that the phase defect PD is covered with the absorber ABS. Therefore, it can be seen that the mask M in the fourth embodiment can eliminate the influence of the phase defect PD even if the phase defect PD exists.

  The mask manufacturing method according to the fourth embodiment will be described below with reference to the drawings. FIG. 20 is a flowchart for explaining a mask manufacturing method according to the fourth embodiment. First, a mask blank having a multilayer film is prepared (S301). Then, the prepared mask blank is inspected for phase defects (S302). For this inspection step, for example, the inspection method described in the first to third embodiments can be used. Thereafter, it is determined whether or not there is a defect on the entire mask blank (S303). If the phase defect does not exist in the mask blank as a result of the inspection described above, an absorber is formed on the mask blank and the absorber is patterned. Thus, a mask is manufactured (S305). On the other hand, if a phase defect exists in the mask blank as a result of the inspection described above, an absorber is formed on the mask blank, and the phase defect is covered by the absorber when patterning is performed on the absorber. The absorber is patterned (S304). In this way, a mask can be manufactured.

  More specifically, when a phase defect is detected in the above-described inspection process, position information of the phase defect present in the mask blank is stored in a memory. At this time, the position information of the phase defect can be accurately grasped by using an alignment mark such as a step pattern provided in advance on the mask blank as a reference.

  And an absorber is formed on a mask blank and this absorber is patterned. At this time, an absorber pattern mask for defining a position where the absorber pattern is formed is used. That is, after the absorber is formed on the entire surface of the mask blank, the absorber formed on the mask blank is patterned by the lithography technique using the absorber pattern mask. Here, the relative positional relationship between the mask blank and the absorber pattern mask is determined based on the stored position information of the phase defect. That is, based on the stored position information of the phase defect, the relative positional relationship between the mask blank and the absorber pattern mask is determined so that the phase defect is covered by the absorber. Thereby, the absorber pattern can be positioned so as to cover the phase defect existing in the mask blank with the absorber. Then, an absorber pattern is formed on the mask blank based on the determined relative position of the mask blank and the absorber pattern mask. As a result, in the manufactured mask, since the phase defect is hidden under the absorber, there is no problem in the exposure projection of the mask pattern onto the semiconductor substrate (semiconductor wafer).

  In order to form the absorber pattern, after the inspection of the mask blank is completed, the absorber material is first uniformly formed on the mask blank. Then, a mask pattern drawing method using normal electron beam lithography is employed. Even if the absorber material is uniformly formed on the mask blank, alignment marks such as step patterns (recesses) formed on the mask blank appear on the surface of the absorber material, so the alignment mark is detected by an electron beam. It is possible to do. Therefore, the alignment mark formed on the mask blank can also be used when forming the absorber pattern. As a result, an absorber pattern that is not affected by the phase defect as in the fourth embodiment can be formed.

  The reference alignment mark is formed by, for example, forming a multilayer film on a mask substrate (mask blank) and then irradiating the multilayer film with FIB (Focused Ion Beam) or a short wavelength laser. be able to. A method of optically detecting the edge position of the mask blank can also be employed.

  In addition, if the phase defect on the mask blank is a small one that does not affect the transfer by itself, if there is no absorber in the vicinity of the phase defect, the main cause of the dimensional variation of the projected exposure pattern It will not be. For this reason, for example, as shown in FIG. 21, when the phase defect PD is fine (smaller than the line and space of the absorber pattern), the absorber pattern made of the absorber ABS formed on the surface of the multilayer film. The absorber pattern is positioned so as to be sufficiently separated from the phase defect PD (a distance larger than the line and space of the absorber pattern), and the absorber pattern is formed on the mask blank based on the positioning result. Can do. In the mask M obtained in this way, there is no phase defect PD in the vicinity of the absorber pattern, so that pattern transfer is performed without hindering exposure projection of the absorber pattern onto the semiconductor substrate (semiconductor wafer). be able to.

  The mask manufacturing method described above is a method of substantially removing the influence of phase defects by adjusting the arrangement position of the entire absorber pattern prepared in advance on the mask blank. Furthermore, for example, in the region where the pattern density is relatively small in the entire region of the absorber pattern, the shape of the absorber pattern is partially changed within a range that does not affect the performance of the finally completed semiconductor device. Thus, the local absorber pattern is redesigned so that the distance between the absorber pattern to be transferred and the phase defect is a predetermined distance (for example, the distance between the line and space of the absorber pattern) or more. You may comprise.

  As described above, according to the mask manufacturing method of the fourth embodiment, the position of the phase defect can be specified, the absorber pattern for forming the semiconductor device, the phase defect present in the mask blank, Can be adjusted so as not to be affected by pattern transfer due to phase defects. As a result, the frequency with which a mask blank having a phase defect can be used as a non-defective product is increased. For this reason, the yield of the mask blank can be greatly improved, and the cost of the mask to be manufactured can be reduced.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  The present invention can be widely used in the manufacturing industry for manufacturing semiconductor devices.

1 CPU
1S semiconductor substrate 2 memory 11 inspection light irradiation region 12 inspection light irradiation region 13 enlarged inspection image 14 enlarged inspection image 15 mirror 16 imaging surface 17 imaging surface 20 inspection light irradiation region 21 enlarged inspection image 22 imaging surface 23 imaging surface ABS absorber AP1 stop AP2 stop BM1 DUV light BM2 DUV light BSP beam splitter BUF buffer layer CAL concave lens CAP capping layer CNT control circuit CF conductor film CVL convex lens CVL1 convex lens CVL2 convex lens CFM3 convex lens HIM pattern Magnified inspection image IMA2 Magnified inspection image LOS1 Illumination optical system LOS2 Imaging optical system LS Light source L1 Lens L2 Lens L3 Lens L4 Lens M Mask MA Alignment mark MB Mask mask Link MDE Device pattern area MLF Multilayer film MS Mask substrate OBL Objective lens PD Phase defect PD1 Phase defect PD2 Phase defect PIX1 Pixel PIX2 Pixel PRA Projection area SE Image detector SE1 First image detector SE2 Second image detector SP1 Inspection stripe ST stage

Claims (8)

(A) placing a mask blank on which a multilayer film for reflecting EUV light is formed on a stage;
(B) after the step (a), irradiating the mask blank with inspection light;
(C) After the step (b), when collecting an enlarged inspection image of the inspection light irradiation region through the objective lens while scanning the inspection light irradiation region of the mask blank, Collecting the enlarged inspection image at a predetermined amount of defocus position;
(D) After the step (b), while collecting the enlarged inspection image of the inspection light irradiation region through the objective lens while scanning the inspection light irradiation region of the mask blank, Collecting the enlarged inspection image at a defocus position of a second predetermined amount different in sign from the first predetermined amount;
(E) determining the presence or absence of a defect based on the enlarged inspection image collected in the step (c) and the enlarged inspection image collected in the step (d);
(F) A mask blank inspection method comprising: a step of displaying the determination result obtained in the step (e). The mask blank inspection method according to claim 1, further comprising:
(G) After the step (b) and before the step (c) and the step (d), the objective lens is reflected from the different first irradiation region and second irradiation region in the inspection light irradiation region, respectively. Branching a first light beam corresponding to the first irradiation region to be transmitted and a second light beam corresponding to the second irradiation region;
In the step (c), using the first light beam branched in the step (g), a first imaging surface of the objective lens that forms a first enlarged inspection image of the first irradiation region or the first imaging surface thereof. Collecting the first enlarged inspection image at a position having an optical path length shorter than one conjugate plane;
In the step (d), the second imaging surface of the objective lens that forms a second enlarged inspection image of the second irradiation region using the second light flux branched in the step (g) or the second imaging surface thereof. 2. A mask blank inspection method comprising collecting the second enlarged inspection image at a position having an optical path length longer than two conjugate planes. An inspection method for a mask blank according to claim 2,
In the step (c), the first enlarged inspection image is obtained by moving the first photodetector along the inspection optical axis in a direction closer to the objective lens than the first imaging plane or the first conjugate plane. Collect and
In the step (d), the second enlarged inspection image is obtained by moving the second photodetector in the direction away from the objective lens along the inspection optical axis with respect to the second imaging surface or the second conjugate surface. A method for inspecting a mask blank, characterized in that An inspection method for a mask blank according to claim 2,
In the step (c), a concave lens is inserted between the objective lens and the first photodetector so that the first imaging plane or the first conjugate plane can be moved from the light receiving position of the first photodetector. Collecting the first enlarged inspection image in a state shifted to a position far from the objective lens,
In the step (d), by inserting a convex lens between the objective lens and the second photodetector, the second imaging plane or the second conjugate plane is moved from the light receiving position of the second photodetector. And the second enlarged inspection image is collected in a state shifted to a position close to the objective lens. An inspection method for a mask blank according to claim 1,
In the step (b), DUV light having a wavelength of 180 nm or more and 210 nm or less is used as the inspection light. An inspection method for a mask blank according to claim 1,
The step (e)
(E1) The enlarged inspection image corresponding to a predetermined area of the mask blank among the enlarged inspection images collected in the step (c), and the mask among the enlarged inspection images collected in the step (d). Calculating a light intensity difference with the enlarged inspection image corresponding to the predetermined area of the blank;
(E2) A method for inspecting a mask blank, comprising the step of determining that a defect exists when the absolute value of the light intensity difference exceeds a preset threshold value after the step (e1). An inspection method for a mask blank according to claim 1,
The step (e) further comprises discriminating between the defect having a convex surface shape and the defect having a concave surface shape. (A) a step of inspecting the presence or absence of defects present in the mask blank on which a multilayer film that reflects EUV light is formed;
(B) when the defect is detected in the step (a), forming an absorber pattern based on the position coordinates of the detected defect,
The step (a)
(A1) placing the mask blank on a stage;
(A2) After the step (a1), a step of irradiating the mask blank with inspection light;
(A3) After the step (a2), when collecting an enlarged inspection image of the inspection light irradiation region through the objective lens while scanning the inspection light irradiation region of the mask blank, Collecting the enlarged inspection image at a predetermined amount of defocus position;
(A4) After the step (a2), when collecting the enlarged inspection image of the inspection light irradiation region via the objective lens while scanning the inspection light irradiation region of the mask blank, Collecting the enlarged inspection image at a defocus position of a second predetermined amount different in sign from the first predetermined amount;
(A5) determining whether or not there is a defect based on the enlarged inspection image collected in the step (a3) and the enlarged inspection image collected in the step (a4);
(A6) A method of manufacturing a mask, comprising the step of storing position coordinates of the defect when the defect present in the mask blank is detected as a result of performing the step (a5).

Images