JP4819191B2 - Mask blank substrate, mask blank, photomask, and semiconductor device manufacturing method - Google Patents

Description

  The present invention relates to a method for manufacturing a mask blank substrate used for a mask blank for producing a photomask used in a photolithography process.

  A photomask is used in a photolithography process of a semiconductor manufacturing process. As the miniaturization of semiconductor devices progresses, the demand for miniaturization in this photolithography process is increasing. In particular, the exposure apparatus using ArF exposure light (193 nm) has been increased in NA in order to cope with miniaturization, and further increased in NA due to the introduction of immersion exposure technology. In order to cope with such a demand for high miniaturization and high NA, it is required to increase the flatness of the photomask. That is, as the pattern line width becomes finer, the width in which the displacement of the transferred pattern due to the flatness is allowed is reduced, and as the NA increases, the depth of focus in the photolithography process is increased. Therefore, the flatness of the main surface of the mask substrate, particularly on the pattern forming side (hereinafter, the main surface on this side is simply referred to as the main surface or the substrate main surface) cannot be ignored.

  FIG. 6 is a view showing the shape of the substrate of the photomask before chucking (before suction) and after chucking (after suction) to the exposure apparatus, and FIG. It is a figure which shows a shape, FIG.6 (b) is a figure which shows the shape of the board | substrate after adsorption | suction. As can be seen from FIG. 6A, the four corner portions of the substrate are slightly higher than the height of the main surface of the chuck area, and gradually increase toward the center. That is, a substantially circular contour line is shown on the substrate before adsorption. As can be seen from FIG. 6 (b), the substrate after the adsorption shows substantially rectangular contour lines. As described above, when the photomask is chucked on the mask stage of the exposure apparatus by the vacuum chuck, the photomask may be greatly deformed at the time of chucking due to the compatibility with the mask stage and the vacuum chuck.

  Conventionally, product management is performed based on the flatness of the photomask before chucking, so even if the main surface shape before chucking is a good product with high flatness, exposure depends on the compatibility with the mask stage and vacuum chuck. When chucked on the mask stage of the apparatus, it may be deformed, and the flatness of the photomask may be greatly deteriorated. In particular, the tendency is remarkable in the substrate having a low symmetry of the shape of the main surface and a twisted shape. For this reason, it is necessary to consider the flatness when the photomask is chucked on the vacuum chuck. Conventionally, a method for selecting a mask substrate with good flatness after being chucked on a mask stage of an exposure apparatus has been proposed (for example, see Patent Document 1).

JP 2003-50458 A

  By using the mask substrate selection method disclosed in Patent Document 1, it becomes possible to relatively easily select a mask substrate having a flatness after chucking of a predetermined value or more. However, with the miniaturization of the transfer pattern, the flatness condition required for the shape of the mask substrate after chucking has become increasingly severe. For example, in the case of a mask substrate (mask blank substrate) having a size of 152 mm square, the flatness in a 132 mm square region is required to be as high as 0.16 μm, and further 0.08 μm. When a substrate having such flatness is selected by the selection method of Patent Document 1, the ratio of acceptable products is greatly reduced, and there is a problem that the production yield deteriorates.

  On the other hand, with regard to the problems related to the deformation of the photomask substrate when it is chucked on the mask stage, diligent research has been carried out on the supply side of the exposure apparatus. As a result, an exposure apparatus having a function of correcting the height direction (cross-sectional direction of the substrate) corresponding to the shape of the photomask during exposure has been developed. In the case of a photomask used in an exposure apparatus capable of correcting the height of the photomask shape when chucked on such a mask stage, there is room for relaxation than the conventional selection criteria for mask blank substrates. . If the selection criteria can be relaxed, the production yield of mask blank substrates can be improved. A method for selecting a mask blank substrate corresponding to an exposure apparatus having such a correction function is desired.

  The present invention has been made in view of this point, and an object of the present invention is to provide a mask blank substrate manufacturing method corresponding to an exposure apparatus having a photomask shape height direction correction function.

  The mask blank substrate manufacturing method of the present invention is a mask blank substrate manufacturing method used for a photomask chucked by a mask stage of an exposure apparatus, and a step of preparing a substrate whose main surface is precisely polished; A step of measuring a pre-chuck main surface shape in an actual measurement region on the main surface, and a photomask produced from the substrate based on the pre-chuck main surface shape of the substrate and the shape of the mask stage. A step of obtaining a post-chuck main surface shape of the substrate when set to a simulation, a step of selecting a substrate having a flatness within a virtual calculation region of the post-chuck main surface shape of a first threshold value or less, A first approximate curve that approximates the cross-sectional shape along the first direction within the correction region of the main surface shape after chucking for the selected substrate. Calculating an approximate curved surface from the first approximate curve and performing subtraction correction from the post-chuck main surface shape to calculate a corrected main surface shape; and within a correction region of the post-correction main surface shape Selecting the one whose flatness is less than or equal to the second threshold value.

  According to this method, the main surface shape after chucking of the substrate (photomask) when the photomask is chucked by the mask stage of the exposure apparatus is obtained by simulation, and then the height direction ( The flatness of the photomask can be calculated by performing correction similar to the correction of the main surface shape of the substrate in the (cross-sectional direction of the substrate). As a result, in the case where the correction in the height direction of the exposure apparatus is not taken into consideration, from the rejected mask blank substrates that could not satisfy the standard flatness, the flatness after correction satisfies the standard flatness. The ratio of products will improve. Therefore, a significant improvement in production yield can be achieved.

  The mask blank substrate manufacturing method of the present invention is a mask blank substrate manufacturing method used for a photomask chucked by a mask stage of an exposure apparatus, and a step of preparing a substrate whose main surface is precisely polished; A step of measuring a pre-chuck main surface shape in an actual measurement region on the main surface, and a photomask produced from the substrate based on the pre-chuck main surface shape of the substrate and the shape of the mask stage. A step of obtaining a post-chuck main surface shape of the substrate when set to a simulation, a step of selecting a substrate having a flatness within a virtual calculation region of the post-chuck main surface shape of a first threshold value or less, A first approximate curve that approximates the cross-sectional shape along the first direction within the correction region of the main surface shape after chucking for the selected substrate. , Calculate a second approximate curve that approximates a cross-sectional shape along a second direction orthogonal to the first direction, calculate an approximate curved surface from the first approximate curve and the second approximate curve, and perform post-chucking Performing subtraction correction from the main surface shape, calculating a corrected main surface shape, and selecting a flatness within the correction region of the corrected main surface shape that is equal to or less than a second threshold value. It is characterized by having.

  According to this method, when the correction in the height direction of the exposure apparatus functions not only in one direction (first direction) but also in a direction orthogonal to the one direction (second direction), exposure is performed. When the correction in the height direction of the device is not taken into account, the ratio of rejected mask blank substrates that could not satisfy the standard flatness to the acceptable products that satisfy the standard flatness in the corrected flatness Is further improved.

  In the mask blank substrate manufacturing method of the present invention, the first approximate curve is preferably a quadratic curve or a quartic curve.

  In the mask blank substrate manufacturing method of the present invention, the second approximate curve is preferably a quadratic curve or a quartic curve.

  When the photomask is chucked on the mask stage of the exposure apparatus, the substrate tends to be deformed into a quadric surface by the chucking force. In the step of polishing the substrate for the photomask blank, the shape of the main surface of the substrate is processed aiming at a convex shape that is relatively high in the central portion and relatively low in the peripheral portion in consideration of this deformation. However, there are many cases in which a substrate having a main surface shape having a large secondary component is formed due to variations in polishing processing accuracy and processing to avoid concave shapes, and the photomask is chucked on the mask stage. In many cases, the secondary component remains in the post-chuck main surface shape of the substrate after being processed, and the flatness does not satisfy a predetermined value. By applying a quadratic curve to the first approximate curve and the second approximate curve and correcting the secondary component of the main surface shape after chucking, the flatness of a predetermined value or more can be obtained, and the substrate that is an acceptable product can be obtained. The ratio is further improved.

  Further, in the case of a substrate having a main surface shape before chucking that has a strong quaternary component (a quaternary curved surface is strong), deformation of the tendency of the quadratic curved surface is added when the photomask is chucked on the mask stage. The main surface after chucking tends to have a shape in which a quaternary component remains (prone to a quartic curved surface). In such a case, by applying a quartic curve to the first approximate curve or the second approximate curve and correcting the quaternary component of the post-chuck main surface shape, the flatness of a predetermined value or more can be obtained, The ratio of substrates that are acceptable products is further improved.

  In the method for manufacturing a mask blank substrate according to the present invention, it is preferable that the first threshold value is 0.32 μm and the second threshold value is 0.16 μm.

  In the mask blank substrate manufacturing method of the present invention, it is more preferable that the first threshold value is 0.24 μm and the second threshold value is 0.08 μm.

  If the optical correction of the exposure apparatus such that the flatness correction amount of the main surface of the substrate exceeds 0.16 μm by the first approximate curve or the second approximate curve is allowed, the transfer pattern on the photomask is actually transferred onto the semiconductor wafer by the exposure apparatus. This is because the astigmatism at the time of transfer to the resist film becomes large, and the image formation of the transfer pattern may be deteriorated and the pattern resolution exceeding a predetermined level may not be satisfied.

  In the mask blank substrate manufacturing method of the present invention, it is preferable that the actual measurement region is a region including the virtual calculation region and the correction region, and the virtual calculation region is a region including the correction region.

  In the mask blank substrate manufacturing method of the present invention, the virtual calculation area is more preferably a 142 mm square area.

  Since the deformation of the substrate is caused by the exposure apparatus chucking the photomask, it is necessary to consider the region outside the region where the transfer pattern is formed. The area where the transfer pattern is formed on the thin film of the photomask is generally inside 132 mm × 104 mm, and there is often no problem if the flatness of the area within the 132 mm square is good. If the flatness is poor, the amount of substrate deformation before and after chucking may be large. If the deformation amount of the substrate is large, the movement amount of the transfer pattern formed on the main surface of the substrate is large, and the pattern position accuracy is lowered. Considering these, it is preferable that the virtual calculation area when selecting the substrate to be simulated is an area within 142 mm square.

  In the mask blank substrate manufacturing method of the present invention, the correction region is more preferably a 132 mm square region.

  In general, the area where the transfer pattern is formed on the thin film of the photomask is generally inside 132 mm × 104 mm. This is because it is preferable to set a correction area in the above-described area so as to ensure that the flatness of the main surface of the substrate after chucking is within a predetermined value so that the transfer pattern may be formed in any direction.

  The method for manufacturing a mask blank substrate according to the present invention preferably includes a step of selecting a substrate having a flatness of 0.4 μm or less in the actual calculation region of the pre-chuck main surface shape.

  In the mask blank substrate manufacturing method of the present invention, the actual calculation area is preferably an area including the virtual calculation area and the correction area.

  In the mask blank substrate manufacturing method of the present invention, the actual calculation region is more preferably a region within 142 mm square.

  If the main surface shape after chucking of the substrate satisfies the first threshold value, the flatness of the main surface shape before chucking may not be deteriorated. A substrate having a good post-chuck main surface shape and a poor flatness of the pre-chuck main surface shape has a large substrate deformation amount before and after the chuck. As a result, the amount of movement of the transfer pattern formed on the main surface of the substrate also increases, and the pattern position accuracy decreases. A substrate having a flatness within the actual calculation region of 0.4 μm or less may be selected as the substrate for simulation. The actual calculation region is preferably a region including a virtual calculation region that is a region for calculating flatness from the post-chuck main surface shape after simulation and a correction region that is a region for performing correction for subtracting the approximate curved surface. . Furthermore, considering the measurement accuracy of the apparatus for measuring flatness, the accuracy of simulation, and the like, the actual calculation region is more preferably a region within 142 mm square.

  In the mask blank substrate manufacturing method of the present invention, for an exposure apparatus that irradiates the photomask with exposure light through a slit that is movable in the first direction and extends in the second direction. More preferred.

  The mask blank manufacturing method of the present invention is characterized in that a thin film is formed on the main surface on the side where the pre-chuck main surface shape of the mask blank substrate obtained by the above method is measured.

  The photomask manufacturing method of the present invention is characterized in that a transfer pattern is formed on a thin film of a mask blank obtained by the above method.

  The mask blank substrate manufacturing method of the present invention is a mask blank substrate manufacturing method used for a photomask chucked by a mask stage of an exposure apparatus, and is a main surface before chucking in an actual measurement region on the main surface of the substrate. Measuring the shape, and the main surface shape after chucking of the substrate when a photomask manufactured from the substrate is set in the exposure apparatus based on the main surface shape before chucking of the substrate and the shape of the mask stage And calculating a first approximate curve that approximates a cross-sectional shape along the first direction within the correction region of the main surface shape after chucking for the substrate, and calculating an approximate curved surface from the first approximate curve Performing a correction to be subtracted from the post-chuck main surface shape, calculating the corrected main surface shape, and the post-correction main surface shape. Flatness correction within the region of the surface shape, and a step of selecting not more than a predetermined threshold value. According to this method, the main surface shape after chucking of the substrate (photomask) when the photomask is chucked by the mask stage of the exposure apparatus is obtained by simulation, and then the height direction ( The flatness of the photomask can be calculated by performing correction similar to the correction of the main surface shape of the substrate in the (cross-sectional direction of the substrate). As a result, in the case where the correction in the height direction of the exposure apparatus is not taken into consideration, from the rejected mask blank substrates that could not satisfy the standard flatness, the flatness after correction satisfies the standard flatness. The ratio of products will improve. Therefore, a significant improvement in production yield can be achieved.

It is a figure for demonstrating a part of exposure apparatus provided with the correction | amendment function of the photomask shape, (a) is a top view, (b) is a side view. It is a flowchart for demonstrating the manufacturing method of the board | substrate for mask blanks concerning embodiment of this invention. It is a figure for demonstrating the main surface shape correction | amendment in a scanning direction, (a) is a figure which shows the position which acquires the cross-sectional shape of a board | substrate, (b) is a figure which shows the cross-sectional shape of a board | substrate. It is a figure for demonstrating the main surface shape correction | amendment in a slit direction, (a) is a figure which shows the position which acquires the cross-sectional shape of a board | substrate, (b) is a figure which shows the cross-sectional shape of a board | substrate. It is a figure which shows schematic structure of the sputtering device used when manufacturing the mask blank which concerns on embodiment of this invention. It is a figure which shows the shape of the board | substrate of the photomask before and after chucking a board | substrate to an exposure apparatus, (a) is a figure which shows the chuck | zipper main surface shape, (b) is a chuck | zipper main surface It is a figure which shows a shape. It is a figure for demonstrating the main surface shape correction | amendment in Example 1, (a) is a figure which shows the position which acquires the cross-sectional shape of a board | substrate, (b) is a figure which shows the cross-sectional shape of a board | substrate, (c) is a figure which shows an approximate curved surface, (d) is a figure which shows the main surface shape after correction | amendment. It is a figure for demonstrating the main surface shape correction | amendment in Example 2, (a) is a figure which shows the position which acquires the cross-sectional shape of a board | substrate, (b) is a figure which shows the cross-sectional shape of a board | substrate, (c) is a figure which shows an approximate curved surface, (d) is a figure which shows the main surface shape after correction | amendment. It is a figure for demonstrating the main surface shape correction | amendment in Example 3, (a) is a figure which shows an approximate curved surface, (b) is a figure which shows the main surface shape after correction | amendment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The mask blank substrate manufacturing method of the present invention is to obtain a photomask mask blank substrate that can be used in an exposure apparatus having a photomask shape correction function. Here, an exposure apparatus having a photomask shape correction function will be described.

  FIG. 1 is a view for explaining a part of an exposure apparatus having a photomask shape correction function, wherein (a) is a plan view and (b) is a side view. In this exposure apparatus, a photomask 2 is placed on a mask stage 1, and the photomask 2 is chucked to the mask stage 1 by a chuck 1a. Above the mask stage 1, an illumination optical system 5 and a slit member 3 having a slit 3a are disposed, and a light source 4 is disposed above the slit member 3. Also, below the mask stage 1, the reduction optical system 6 and the semiconductor wafer W placed on the wafer stage 7 are located.

  In such an exposure apparatus, the exposure to the semiconductor wafer W is performed while moving the mask stage 1 chucking the photomask 2 in the scanning direction and moving the wafer stage 7 in the direction opposite to the moving direction of the mask stage 1. Done. At this time, the light from the light source 4 is irradiated to the photomask 2 chucked by the mask stage 1 through the slit 3 a, and irradiated to the semiconductor wafer W through the photomask 2. As a result, the transfer pattern is exposed to the photoresist provided on the semiconductor wafer W. The moving direction (scanning direction) of the mask stage 1 and the extending direction (longitudinal direction) of the slit 3a are substantially orthogonal to each other.

  In this exposure apparatus, main surface shape correction can be performed in accordance with the shape of a photomask obtained by measurement in advance. In the main surface shape correction, in the scanning direction, the main surface shape correction is performed by changing the scan trajectory by changing the relative distance between the mask stage and the wafer stage 7 on which the semiconductor wafer W is placed. On the other hand, in the slit direction, main surface shape correction may be performed by changing the shape of illumination light by changing astigmatism. In the case of this exposure apparatus, a description has been given of a type in which the main surface shape correction can be corrected in two directions of the scan direction and the slit direction, but depending on the exposure apparatus, when the main surface shape correction is only in the scan direction, Some are only in the slit direction.

  In the mask blank substrate manufacturing method of the present invention, a mask blank substrate manufacturing method used for a photomask chucked by a mask stage of an exposure apparatus, the step of preparing a substrate whose main surface is precisely polished And a step of measuring a pre-chuck main surface shape in an actual measurement area on the main surface, and exposing the photomask produced from the substrate based on the pre-chuck main surface shape of the substrate and the shape of the mask stage. A step of obtaining a post-chuck main surface shape of the substrate when set in an apparatus by simulation, a step of selecting a substrate having a flatness within a virtual calculation region of the post-chuck main surface shape of a first threshold value or less, The selected substrate is approximated to a cross-sectional shape along the first direction within the correction region of the post-chuck main surface shape. Calculating an approximated curve, calculating an approximated curved surface from the first approximated curve, subtracting from the post-chuck main surface shape, and calculating a corrected main surface shape; and a correction region of the corrected main surface shape And a step of selecting one having a flatness equal to or less than a second threshold value.

  Alternatively, in the mask blank substrate manufacturing method of the present invention, a mask blank substrate manufacturing method used for a photomask chucked by a mask stage of an exposure apparatus, the main surface of which is precisely polished is prepared. A step of measuring a pre-chuck main surface shape in an actual measurement region on the main surface, and a photomask fabricated from the substrate based on the pre-chuck main surface shape of the substrate and the shape of the mask stage. A step of obtaining a post-chuck main surface shape of the substrate when set in the exposure apparatus by simulation, and a step of selecting a substrate having a flatness within a virtual calculation region of the post-chuck main surface shape of a first threshold value or less. And a cross-sectional shape along the first direction within the correction region of the main surface shape after chucking for the selected substrate. A similar first approximate curve is calculated, a second approximate curve that approximates a cross-sectional shape along a second direction orthogonal to the first direction is calculated, and an approximate curved surface is calculated from the first approximate curve and the second approximate curve. A step of calculating and subtracting from the post-chuck main surface shape to calculate a post-correction main surface shape, and a flatness within a correction region of the post-correction main surface shape being a second threshold value or less And a step of selecting.

  FIG. 2 is a flowchart for explaining a method for manufacturing a mask blank substrate according to the embodiment of the present invention. In this manufacturing method, first, a mask blank substrate whose main surface is precisely polished is manufactured (ST11).

  In the present invention, a glass substrate can be used as the mask blank substrate. The glass substrate is not particularly limited as long as it is used as a mask blank. Examples thereof include synthetic quartz glass, soda lime glass, aluminosilicate glass, borosilicate glass, and alkali-free glass.

  Such a mask blank substrate can be manufactured through, for example, a rough polishing process, a precision polishing process, and an ultraprecision polishing process.

  Next, the height information in the measured region on the main surface of the mask blank substrate is acquired, and from this height information, the shape of the main surface before chucking in the sectional view of the mask blank substrate, that is, information on the main surface shape before chucking Is acquired (ST12). The height information here refers to height information from a reference surface at a plurality of measurement points in an actual measurement region provided in the main surface of the mask blank substrate. For example, when the size of the mask blank is 152 mm × 152 mm, the actually measured region can be an inner region of 146 mm × 146 mm. The actual measurement area is a wide area including at least a virtual calculation area and a correction area described later.

  The main surface shape before chucking of the mask blank substrate is obtained by measuring with a wavelength shift interferometer using a wavelength modulation laser. This wavelength shift interferometer calculates the difference in height of the measured surface from the interference fringes between the reflected light reflected from the measured surface and the back surface of the mask blank substrate and the measuring machine reference surface (front reference surface). The frequency difference of each interference fringe is detected, and the interference fringe with the measuring machine reference surface (front reference surface) by the reflected light reflected from the measured surface and the back surface of the mask blank substrate is separated, The uneven shape of the measurement surface is measured.

  Further, in order to perform a simulation described later with high accuracy, it is desirable to increase as many measurement points as possible to obtain height information. However, although more accurate simulation results can be obtained when the number of measurement points is increased, it takes time for the simulation. Therefore, it is preferable to determine the measurement points in consideration of these points. For example, the measurement points can be 256 × 256 points.

  Next, the flatness of the mask blank substrate is calculated from the difference between the maximum value and the minimum value in the actual calculation region including the transfer region of the photomask, based on the main surface shape before chucking of the substrate obtained by measurement (ST13). ). Further, it is determined whether or not the flatness obtained in this way is below an allowable value (ST14). A mask blank substrate having a flatness larger than an allowable value is rejected and is not supplied to a subsequent process. Even if the flatness of the main surface shape after chucking of the mask blank substrate photomask manufactured by this manufacturing method is good, there may be a problem if the flatness of the main surface shape before chucking is poor. A substrate having a large flatness change amount before and after chucking has a large substrate deformation amount. A photomask manufactured from a substrate with a large amount of substrate deformation may increase the amount of movement of the transfer pattern formed on the main surface of the substrate before and after chucking, which may reduce the pattern position accuracy after chucking. is there. Considering this point, select a substrate whose flatness before chucking is less than an allowable value. If the flatness tolerance before chucking is too high, the production yield deteriorates and the purpose of the present application cannot be achieved. If it is too low, a photomask may be produced from a substrate having a large substrate deformation amount. . In consideration of these balances, the allowable value of the flatness of the substrate before chucking is set to 0.4 μm or less. Further, in the case of manufacturing a mask blank substrate used for a photomask that requires high pattern position accuracy, it is necessary that the substrate deformation amount be smaller, so that the allowable value of the flatness of the substrate is reduced to 0. .. 3 μm or less.

  On the other hand, the amount of deformation of the substrate before and after chucking also varies depending on the size of the actual calculation area which is an area for calculating the flatness. The larger the actual calculation area, the smaller the deformation amount of the substrate before and after chucking. First, the actual calculation area for calculating the flatness before chucking is required to be an area that is the same as or narrower than the actual measurement area, which is the measurement area, and wider than the virtual calculation area. The predetermined area including the transfer area of the photomask is determined by the exposure wavelength, the type of fine pattern (circuit pattern) formed on the semiconductor wafer, and the like. When the size of the mask blank is 152 mm square, the transfer area of the photomask is often an inner area of 104 mm × 132 mm. In consideration of this, the virtual calculation area can be a square area inside the 132 mm square. However, when the flatness of the outer region is poor, there is a high possibility that the substrate is largely deformed before and after the chuck. Considering this, when the size of the mask blank (photomask) is 152 mm square, the actual calculation area is preferably at least an inner area of 142 mm square. Furthermore, in the case where the measurement accuracy of the main surface shape before chucking and the accuracy of simulation are high, it is preferable to set the inner region of 146 mm square or the inner region of 148 mm square.

  Next, based on the obtained pre-chuck main surface shape and the shape of the mask stage, height information when a photomask manufactured from this substrate is set in an exposure apparatus is obtained by simulation, and this height information is obtained. The main surface shape after chucking in the sectional view of the mask blank substrate is obtained (ST15). In this simulation process, a state in which the photomask is set on the mask stage of the exposure apparatus is simulated to obtain height information from the reference plane at a plurality of measurement points on the main surface of the substrate. Here, when the photomask is set in the exposure apparatus, the information necessary for obtaining the height information of a plurality of measurement points on the substrate by simulating is obtained when acquiring the information on the main surface shape before chucking. The obtained height information from the reference surface of a plurality of measurement points on the main surface of the substrate and the shape information of the mask stage including the region where the mask stage of the exposure apparatus abuts on the main surface of the substrate. Using this information, the height information from the reference plane at multiple measurement points on the main surface of the substrate when a photomask is set on the mask stage of the exposure apparatus is simulated using the deflection differential equation in material mechanics. Can be obtained. And the main surface shape after chuck | zipper in the cross sectional view of a board | substrate is obtained from the obtained height information.

  In this step, based on the information on the main surface shape before chucking of the substrate, simulation is performed when a photomask manufactured from the substrate is chucked on the mask stage of the exposure apparatus. Since the thin film that forms the transfer pattern formed on the surface of the mask blank substrate in the subsequent process is formed with high accuracy by sputtering, the film thickness change in the main surface direction of this thin film is compared with the flatness of the substrate. It is a minute and negligible range. It can be said that even if a simulation is performed based on the main surface shape before chucking of the mask blank substrate, there is no significant difference until the influence.

The bending differential equation is obtained as follows by taking the positive direction of the Z-axis in the direction of gravity.
H ₂ = H ₁ + B ₁ + B ₂ −H _AB
H ₂ : Height information on the main surface of the substrate after chucking H ₁ : Height information on the main surface of the substrate before chucking B ₁ : Warping of the substrate with the mask stage as a fulcrum (leverage effect)
B ₂ : Deflection due to gravity of substrate (≈maximum value is 0.1 μm at substrate center)
H _AB : Average value of height information of the substrate in the region along the scanning direction where the substrate contacts the mask stage

  The shape information of the mask stage described above includes the mask stage on the main surface of the substrate in addition to the position and region where the mask stage contacts the main surface of the substrate (the region having the width in the slit direction and the width in the scan direction). Information on the flatness of the mask stage in the contacted area (surface) may be included. Furthermore, the simulation method is not limited to the above, and a simulation using a general finite element method may be used.

  Next, based on the post-chuck shape obtained by the simulation, the flatness of the mask blank substrate is calculated from the difference between the maximum value and the minimum value in the virtual calculation region including the photomask transfer region (ST16). This flatness contributes to the formation of a good transfer pattern during pattern transfer using an exposure apparatus. The virtual calculation area including the transfer area of the photomask is determined by the exposure wavelength, the type of fine pattern (circuit pattern) formed on the semiconductor wafer, and the like. When the size of the mask blank is 152 mm square, the mask transfer area is often an inner area of 104 mm × 132 mm. In consideration of this, it can be a square area inside a 132 mm square. However, when the flatness of the outer region is poor, there is a possibility that the substrate is largely deformed before and after the chuck. A photomask manufactured from such a substrate has a large amount of movement of the transfer pattern formed on the main surface of the substrate, and the pattern position accuracy after chucking may be lowered. Considering this, when the size of the mask blank (photomask) is 152 mm square, it is preferable to set it to an inner area of at least 142 mm square. Furthermore, in the case where the measurement accuracy of the main surface shape before chucking and the accuracy of simulation are high, it is preferable to set the inner region of 146 mm square or the inner region of 148 mm square.

  Next, it is determined whether or not the flatness obtained in this way is equal to or less than the first threshold value (ST17). The first threshold value of the flatness is determined on the resist film on the semiconductor wafer by the flatness of the corrected main surface shape required for the photomask manufactured from the mask blank substrate and the height correction by the exposure apparatus. The image of the transfer pattern is selected from an allowable value of the flatness correction amount that can satisfy a predetermined pattern resolution. When the flatness of the main surface after chuck required for the photomask (mask blank substrate) is 0.16 μm or less and the allowable value of the flatness correction amount of the exposure apparatus is 0.16 μm, the first threshold value is The total can be 0.32 μm. When the flatness of the main surface after chuck required for the photomask (mask blank substrate) is 0.08 μm or less and the allowable value of the flatness correction amount of the exposure apparatus is 0.16 μm, the first threshold value is The total thickness can be 0.24 μm.

  When the correction technique of the exposure apparatus is improved and the allowable flatness correction amount is increased, the first threshold value can be increased in accordance with the upper limit of the correction amount. For example, when the flatness correction amount is increased to 0.24 μm, 0.32 μm, and the like, and the flatness required for the main surface after chucking of the photomask is 0.16 μm or less, the first threshold value is It can be increased to 0.40 μm and 0.48 μm. When the flatness required for the main surface of the photomask after chucking is 0.08 μm or less, the first threshold can be increased to 0.32 μm and 0.40 μm.

  Conversely, when the exposure apparatus correction technology does not improve and the exposure apparatus is further increased in NA, the focus tolerance between the reduction optical system and the semiconductor wafer becomes smaller, and the allowable flatness correction amount is smaller. Therefore, it is necessary to reduce the first threshold value. For example, when the flatness correction amount is increased to 0.12 μm, 0.10 μmm, 0.08 μm, etc., and the flatness required for the main surface after chucking of the photomask is 0.16 μm or less, One threshold value needs to be reduced to 0.28 μm, 0.26 μm, and 0.24 μm. In addition, when the flatness required for the main surface of the photomask after chucking is 0.08 μm or less, the first threshold needs to be as small as 0.20 μm, 0.18 μm, and 0.16 μm.

  Next, a first approximate curve that approximates the cross-sectional shape along the first direction is calculated in the correction area of the post-chuck main surface shape, an approximate curved surface is calculated from the first approximate curve, and the post-chuck main surface shape Then, the shape correction is performed by subtracting the approximate curved surface, and the corrected main surface shape is calculated (ST18). Further, when an exposure apparatus of a type capable of correcting the main surface shape in two directions of the scan direction and the slit direction is targeted, a first approximate curve that approximates a cross-sectional shape along the first direction is calculated, A second approximate curve that approximates a cross-sectional shape along a second direction orthogonal to the first direction of the correction region is calculated, an approximate curved surface is calculated from the first approximate curve and the second approximate curve, and chucking Shape correction is performed by subtracting the approximate curved surface from the rear main surface shape, and the corrected main surface shape is calculated.

  This main surface shape correction will be described with reference to FIGS. Here, a case will be described in which the first direction is the scanning direction of the exposure apparatus, the second direction is the slit direction of the exposure apparatus, the first approximate curve is a quartic curve, and the second approximate curve is a quadratic curve. . 3A and 3B are diagrams for explaining main surface shape correction in the scanning direction, in which FIG. 3A is a diagram illustrating a position at which the cross-sectional shape of the substrate is acquired, and FIG. is there. 4A and 4B are diagrams for explaining main surface shape correction in the slit direction. FIG. 4A is a diagram illustrating a position at which the cross-sectional shape of the substrate is acquired, and FIG. 4B is a diagram illustrating the substrate shape. is there.

In the main surface shape correction in the scan direction (first direction), as shown in FIG. 3A, the right end, the center, and the center of the post-chuck main surface shape correction region X of the mask blank substrate parallel to the scan direction are shown. By obtaining the cross-sectional shape of the substrate in the scanning direction from the height information on each straight line Y _{1 at the} left end, and calculating a quartic curve with respect to the cross-sectional shapes of these three places by the least square method, FIG. as shown, the approximate curve (the first approximation curve) in the scanning direction to calculate the Z _1. Next, in the main surface correction in the slit direction (second direction), as shown in FIG. 4A, the upper end parallel to the slit direction of the correction region X of the post-chuck main surface shape of the mask blank substrate, central, we obtain a cross-sectional shape of the slit direction of the substrate from the height information in each linear Y ₂ of the lower end, by calculating the quadratic curve by the least square method to this three positions of the cross-sectional shape, FIG. 4 (b as shown in), calculates the slit direction of the approximate curve (the second approximation curve) Z _2. When intended for exposure apparatus for correcting the scan direction only (first direction) calculates the approximated surface from the first approximation curve Z _1, the correction of the slits only in the direction (second direction) when intended for exposure apparatus for performing calculates the approximated surface from the second approximation curve Z _2, the main surface shape after correction by the approximated surface corrects to subtract from the chuck after shape obtained by the above simulation calculate.

Further, the scanning direction (first direction) when directed to an exposure apparatus for correcting from both the slit direction (second direction) from the first approximation curve Z ₁ and the second approximation curve Z ₂ An approximate curved surface is calculated, and the corrected main surface shape is calculated by performing correction by subtracting the approximate curved surface from the post-chuck shape obtained by the simulation. Alternatively, from the chuck after shape obtained by the above simulation, performs correction of subtracting the approximated surface calculated from the first approximation curve Z _1, further performs a correction of subtracting the approximated surface calculated from the second approximation curve Z _2, The corrected main surface shape is calculated.

  This main surface shape correction is for simulating main surface shape correction performed by an exposure apparatus having a correction function, and the corrected main surface shape obtained thereby is a photomask produced from the simulated substrate. Basically the same as the substrate shape after chucking to the exposure device and correcting the main surface shape (from exposure device mechanical error, flatness measurement error, simulation error, shape change with pellicle sticking, etc.) It is not exactly the same, but does not affect the judgment).

  In addition, although it is the correction | amendment area | region X, the transfer area | region in which the transfer pattern of a photomask is formed is decided by the kind of fine pattern (circuit pattern) formed on an exposure wavelength or a semiconductor wafer. When the size of the mask blank is 152 mm square, the mask transfer area is often an inner area of 104 mm × 132 mm. In consideration of this, the correction region X is preferably a square region inside a 132 mm square. When high accuracy is required by a photomask, the inner area is 142 mm square.

  Here, three substrate cross-sectional shapes in the scan direction (first direction) and the slit direction (second direction) of the correction region X are obtained, but the first approximate curve, the second approximate curved surface, In the case of obtaining higher simulation accuracy of the approximate curved surface calculated from these, it may be calculated from the cross-sectional shape of the substrate at four or more locations. Furthermore, although a quartic curve is used as the first approximate curve and a quadratic curve is used as the second approximate curve here, the present invention is not limited to this. A quadratic curve may be applied to the first approximate curve, and a quartic curve may be applied to the second approximate curve, or both the first approximate curve and the second approximate curve are quadratic curves (in this case, the synthesized approximate curved surface is a quadratic curve). A curved surface) or a quartic curve (in this case, the synthesized approximate curved surface becomes a quartic curved surface). It is optimal to select a correction simulation closest to the main surface shape correction function of the exposure apparatus that actually chucks the photomask.

  Next, based on the corrected main surface shape obtained by the main surface shape correction, the flatness after correction of the mask blank substrate is determined from the difference between the maximum value and the minimum value of the height information in the correction region including the transfer region of the photomask. The degree is calculated, and it is determined whether or not the flatness is equal to or less than the second threshold (ST19).

  The corrected main surface shape is obtained by simulating the main surface shape of the substrate after the photomask (mask blank substrate) is chucked on the mask stage of the exposure apparatus and the main surface shape is corrected by the correction function. That is, this corrected main surface shape satisfies the flatness of a pattern transfer region (a correction region is a region including at least the pattern transfer region) required for a photomask manufactured using the mask blank substrate. If it is, it may be determined that the substrate is an acceptable mask blank. Therefore, it is desirable to select the flatness of the pattern transfer region required for the photomask manufactured using the mask blank as the second threshold here. For example, the correction area is an area within a 132 mm square, and the second threshold value is 0.16 μm. In the case of a mask blank substrate used for a photomask that requires higher accuracy, the correction area is an area within a 132 mm square, and the second threshold value may be 0.08 μm. Only the mask blank substrate determined to conform to the flatness of the second threshold is supplied to the mask blank manufacturing and photomask manufacturing processes described later.

  A mask blank can be formed by forming at least a light-shielding film on the main surface of the mask blank substrate determined to be an acceptable product having a flatness of the second threshold value or less (ST20). Examples of the material constituting the light shielding film include chromium, metal silicide, and tantalum. Further, other films, an antireflection film, a phase shift film, and the like may be appropriately formed depending on the use and configuration of the photomask. As a material of the antireflection film, if it is a chromium-based material, CrO, CrON, CrOCN, etc., if it is a MoSi-based material, MoSiON, MoSiO, MoSiN, MoSiOC, MoSiOCN, if it is a tantalum-based material, TaO, TaON, TaBO, TaBON or the like is preferably used. Further, as a material of the phase shift film, it is preferable to use MSiON, MSiO, MSiN, MSiOC, MSiOCN (M: Mo, W, Ta, Zr, etc.) or the like.

  The light shielding film can be formed by a sputtering method. As the sputtering apparatus, a DC magnetron sputtering apparatus, an RF magnetron sputtering apparatus, or the like can be used. When sputtering the light-shielding film onto the mask blank substrate, it is preferable to form the film by rotating the substrate and disposing the target at a position inclined by a predetermined angle from the rotation axis of the substrate. By such a film formation method, the in-plane variation of the light shielding film can be reduced and the film can be formed uniformly.

  In the case where the substrate is rotated and the film is formed by placing the target at a position inclined by a predetermined angle from the rotation axis of the substrate, the in-plane distribution of the phase angle and the transmittance is determined between the substrate and the target. It also changes depending on the positional relationship. The positional relationship between the target and the substrate will be described with reference to FIG. The offset distance (the distance between the central axis of the substrate and a straight line passing through the center of the target and parallel to the central axis of the substrate) is adjusted by the area where the distribution of the phase angle and the transmittance should be ensured. Generally, when the area where the distribution should be secured is large, the necessary offset distance becomes large. In the form of this example, in order to realize a phase angle distribution within ± 2 ° and a transmittance distribution within ± 0.2% within a 142 mm square substrate, the offset distance needs to be about 200 mm to 350 mm, A preferred offset distance is 240 mm to 280 mm. Although the optimum range of the target-substrate vertical distance (T / S) varies depending on the offset distance, a phase angle distribution within ± 2 ° and a transmittance distribution within ± 0.2% are realized within a 142 mm square substrate. Therefore, the target-substrate vertical distance (T / S) needs to be about 200 mm to 380 mm, and the preferable T / S is 210 mm to 300 mm. The target inclination angle affects the film formation speed. In order to obtain a large film formation speed, the target inclination angle is suitably from 0 ° to 45 °, and the preferred target inclination angle is from 10 ° to 30 °.

  A photomask can be manufactured by patterning at least the light shielding film described above by photolithography and etching to provide a transfer pattern (ST21). Note that the etchant for etching is appropriately changed according to the material of the film to be etched.

  The obtained photomask is set on the mask stage of the exposure apparatus, this photomask is used, and the photomask mask pattern is formed on the resist film formed on the semiconductor wafer by using photolithography with ArF excimer laser as exposure light. By transferring, a desired circuit pattern is formed on the semiconductor wafer, and a semiconductor device is manufactured.

  Next, examples performed for clarifying the effects of the present invention will be described. In the following examples, the case where the mask blank substrate is a glass substrate will be described.

Example 1
A predetermined number of glass substrates (approx. 152 mm x 152 mm x 6.45 mm) subjected to lapping and chamfering on a synthetic quartz glass substrate are set in a double-side polishing machine, and a rough polishing process is performed under the following polishing conditions. went. After the rough polishing step, the glass substrate was subjected to ultrasonic cleaning in order to remove the abrasive grains adhering to the glass substrate. The polishing conditions such as the processing pressure, the number of rotations of the upper and lower surface plates, and the polishing time were appropriately adjusted.
Polishing liquid: Cerium oxide (average particle size 2 μm to 3 μm) + water Polishing pad: Hard polisher (urethane pad)

Next, a predetermined number of sheets were set in a double-side polishing apparatus on the glass substrate after rough polishing, and a precision polishing step was performed under the following polishing conditions. After the precision polishing step, the glass substrate was subjected to ultrasonic cleaning in order to remove abrasive grains adhering to the glass substrate. The polishing conditions such as the processing pressure, the number of rotations of the upper and lower surface plates, and the polishing time were appropriately adjusted. The main surface shape on the side of forming the transfer pattern of the glass substrate after the precision polishing step is polished by adjusting various conditions so that the four corners are convex. This is because in the next ultra-precise polishing process, the four corners of the main surface of the substrate are preferentially polished, and this prevents the edge of the four corners from sagging. The flatness within a 142 mm square can be 0.4 μm or less.
Polishing liquid: Cerium oxide (average particle size 1μm) + water Polishing pad: Soft polisher (suede type)

Subsequently, a predetermined number of sheets were set in a double-side polishing apparatus on the glass substrate after precision polishing, and an ultra-precision polishing process was performed under the following polishing conditions. After the ultraprecision polishing step, the glass substrate was ultrasonically cleaned to remove abrasive grains adhering to the glass substrate. The polishing conditions such as the processing pressure, the number of rotations of the upper and lower surface plates, and the polishing time were appropriately adjusted. This ultra-precision polishing process has a characteristic that four corners are preferentially polished due to the substrate shape being square. The polishing conditions are set so that the flatness within the 142 mm square of the main surface of the substrate does not become larger than 0.4 μm while the surface roughness of the main surface of the substrate is set to a predetermined roughness of 0.4 nm or less. . Thus, a glass substrate according to the present invention was produced (ST11).
Polishing liquid: Colloidal silica (average particle size 100 nm) + water Polishing pad: Super soft polisher (suede type)

  With respect to the main surface of the glass substrate thus obtained, an actual measurement region (150 mm × 150 mm) of the main surface (main surface on which a thin film is formed) of the substrate using a wavelength shift interferometer using a wavelength varying laser. In FIG. 6, information on the main surface shape before chucking (height information from the focal plane (virtual absolute plane) calculated by the least square method) is obtained for each measurement point of 256 × 256 (FIG. 6A). (Refer to the shape) and stored in the computer (ST12). Then, the flatness in the actual calculation area (142 mm × 142 mm) is obtained from the height information of the measured pre-chuck main surface shape in the actual measurement area (ST13), and a substrate having an allowable value (0.4 μm) or less is selected ( ST14). As a result, the number of glass substrates satisfying this condition was 99 out of 100. It should be noted that the surface shape of the main surface of the substrate was such that the height of the main surface gradually decreased from the central region toward the peripheral edge by this height information.

  Next, from the obtained pre-chuck main surface shape information and the mask stage shape information of the exposure apparatus, each measurement point is measured from the reference plane when the substrate is set on the exposure apparatus for each measurement point using the above-described deflection differential equation. The post-chuck main surface shape, which is height information, was calculated by simulation (see FIG. 6B, the post-chuck main surface shape) (ST15). Then, from this simulation result, the difference between the maximum value and the minimum value from the reference surface in the virtual calculation area (142 mm × 142 mm) including the transfer area of the photomask was obtained, and the flatness in this virtual calculation area was calculated ( ST16). And the board | substrate whose flatness is below a 1st threshold value (0.32 micrometer) was selected (ST17). As a result, 98 of the 99 substrates satisfied this condition.

Next, main surface shape correction in the scan direction (first direction) is performed (ST18). The glass substrate having the post-chuck main surface shape shown in FIG. 6 (b), as shown in FIG. 7 (a), the right end parallel to the scanning direction of the correction region X of the post-chuck main surface shape of the mask blank substrate, central obtains the scanning direction of the substrate cross-sectional shape from the height information in each linear Y ₁ at the left end, by calculating the fourth-order curve with the minimum square method for this three positions of the cross-sectional shape, FIG. 7 (b the scanning direction of the approximate curve shown in) (first approximation curve) to calculate the Z _1. Then, to calculate the approximate surface as shown in FIG. 7 (c) from the approximate curve Z _1, subtracting from the chuck after the shape obtained above simulation. It corrected main surface shape of the glass substrate after subtracting the approximated surface Z ₁ shown in FIG. 7 (d).

  Next, the calculated flatness in the correction region (132 mm × 132 mm) of the corrected main surface was calculated, and a substrate having a second threshold value (0.16 μm) or less was selected (ST19). The second threshold value is a standard flatness for which the mask blank substrate is required. As a result, the number of glass substrates satisfying this condition was 96 out of 98. When a substrate satisfying the condition of 0.16 μm or less, which is the same as the second threshold, is selected for the flatness (ST16) obtained from the main surface shape after chucking of the substrate calculated in the conventional simulation (ST15), 90 of 98 substrates are selected. Therefore, it can be seen that the production yield is greatly improved by performing the main surface shape correction (ST18).

Next, a back surface antireflection layer, a light shielding layer, and a surface antireflection layer were formed in that order as a thin film (light shielding film) for forming a transfer pattern on the glass substrate obtained as described above (ST20). . Specifically, a Cr target is used as a sputtering target, and a mixed gas of Ar, CO ₂ , N ₂ , and He is used as a sputtering gas (gas flow ratio Ar: CO ₂ : N ₂ : He = 24: 29: 12: 35). And a gas pressure of 0.2 Pa, a DC power supply of 1.7 kW, and a CrOCN film having a thickness of 39 nm was formed as a back surface antireflection layer. Next, a Cr target is used as a sputtering target, a mixed gas of Ar, NO, and He is used as a sputtering gas (gas flow ratio Ar: NO: He = 27: 18: 55), a gas pressure is 0.1 Pa, and a DC power supply is used. A CrON film having a thickness of 17 nm was formed as a light shielding layer at 1.7 kW. Next, a Cr target is used as a sputtering target, and a mixed gas of Ar, CO ₂ , N ₂ , and He is used as a sputtering gas (gas flow ratio Ar: CO ₂ : N ₂ : He = 21: 37: 11: 31), A CrOCN film having a thickness of 14 nm was formed as a surface antireflection layer with a gas pressure of 0.2 Pa and a DC power supply of 1.8 kW. The back surface antireflection layer, the light shielding layer, and the surface antireflection layer formed under these conditions had very low stress throughout the light shielding film, and the change in the shape of the substrate could be minimized. In this way, a mask blank was manufactured.

  A photomask (binary mask) was produced by patterning the light-shielding film of the mask blank thus obtained into a predetermined pattern (ST21). The obtained photomask was verified with an exposure apparatus capable of at least correcting the main surface shape in the scanning direction. After chucking on the mask stage of the exposure apparatus and correcting the main surface shape in the scanning direction, the resist film on the semiconductor wafer W was exposed to transfer a photomask pattern. As a result of verifying the CD accuracy and pattern position accuracy of the pattern transferred to the resist film, it was confirmed that this photomask can sufficiently cope with the DRAM hp32 nm generation.

(Example 2)
As in Example 1, the steps from ST11 to ST17 were performed, and 98 glass substrates were selected. Next, main surface shape correction (ST18) in the slit direction (second direction) is performed. 6 (b), the upper end parallel to the slit direction of the correction region X of the post-chuck main surface shape of the mask blank substrate, as shown in FIG. 8 (a), central obtains the slit direction of the substrate cross-sectional shape from the height information in each linear Y ₂ of the lower end, by calculating the quadratic curve by the least square method to this three positions of the cross-sectional shape, FIG. 8 (b slit direction of the approximate curve shown in) (second approximation curve) to calculate the Z _2. Then, to calculate the approximate surface as shown from the approximate curve Z ₂ in FIG. 8 (c), subtracted from the chuck after the shape obtained above simulation. The substrate shape after subtracting the approximated surface Z ₁ shown in FIG. 8 (d).

  Next, the calculated flatness in the correction region (132 mm × 132 mm) of the corrected main surface was calculated, and a substrate having a second threshold value (0.16 μm) or less was selected (ST19). The second threshold value is a standard flatness for which the mask blank substrate is required. As a result, the number of glass substrates satisfying this condition was 95 out of 98. When a substrate satisfying the condition of 0.16 μm or less, which is the same as the second threshold, is selected for the flatness obtained from the post chuck main surface shape (ST15) calculated in the conventional simulation (ST14), 90 of 98 substrates are selected. Therefore, it can be seen that the production yield is greatly improved by performing the main surface shape correction (ST18).

Next, on the glass substrate obtained as described above, as a thin film for forming a transfer pattern, a phase shift film and a light shielding film comprising a back surface antireflection layer, a light shielding layer, and a surface antireflection layer are formed. A mask blank was manufactured (ST20). Specifically, using a target of Mo: Si = 10: 90 (atomic% ratio), a mixed gas of Ar, N ₂ , and He is used as a sputtering gas (gas flow ratio Ar: N ₂ : He = 5: 49: 46). And a gas pressure of 0.3 Pa, a DC power source power of 2.8 kW, and a MoSiN film having a thickness of 69 nm as a phase shift film. Next, the substrate on which the phase shift film was formed was heat-treated (annealed) at 250 ° C. for 5 minutes.

Next, a light shielding film composed of a back surface antireflection layer, a light shielding layer, and a front surface antireflection layer was formed. Specifically, first, a Cr target is used as a sputtering target, and a mixed gas of Ar, CO ₂ , N ₂ , and He is used as a sputtering gas (gas flow ratio Ar: CO ₂ : N ₂ : He = 22: 39: 6 33), a gas pressure of 0.2 Pa, a DC power supply of 1.7 kW, and a CrOCN film having a thickness of 30 nm was formed as a back-surface antireflection layer. Next, a Cr target is used as a sputtering target, a mixed gas of Ar and N ₂ is used as a sputtering gas (gas flow ratio Ar: N ₂ = 83: 17), a gas pressure is 0.1 Pa, and a DC power supply is 1.7 kW. Then, a CrN film having a thickness of 4 nm was formed as a light shielding layer. Next, a Cr target is used as a sputtering target, and a mixed gas of Ar, CO ₂ , N ₂ , and He is used as a sputtering gas (gas flow ratio Ar: CO ₂ : N ₂ : He = 21: 37: 11: 31), A CrOCN film having a thickness of 14 nm was formed as a surface antireflection layer with a gas pressure of 0.2 Pa and a DC power supply of 1.8 kW. The back surface antireflection layer, the light shielding layer, and the surface antireflection layer formed under these conditions have very low stress in the entire light shielding film, and the phase shift film also has very low stress. Minimized.

  Further, the light shielding film and the phase shift film of the mask blank were patterned into a predetermined pattern to produce a photomask (phase shift mask) (ST21). The obtained photomask was verified with an exposure apparatus capable of at least correcting the main surface shape in the slit direction. After chucking on the mask stage of the exposure apparatus and correcting the main surface shape in the slit direction, exposure was performed on the resist film of the semiconductor wafer W to transfer the photomask pattern. As a result of verifying the CD accuracy and pattern position accuracy of the pattern transferred to the resist film, it was confirmed that this photomask can sufficiently cope with the DRAM hp32 nm generation.

(Example 3)
As in Example 1, the steps from ST11 to ST17 were performed, and 98 glass substrates were selected. Next, main surface shape correction (ST18) is performed from both the scanning direction (first direction) and the slit direction (second direction). The glass substrate having the post-chuck main surface shape shown in FIG. 6 (b), as shown in FIG. 7 (a), the right end parallel to the scanning direction of the correction region X of the post-chuck main surface shape of the mask blank substrate, central obtains the scanning direction of the substrate cross-sectional shape from the height information in each linear Y ₁ at the left end, by calculating the fourth-order curve with the minimum square method for this three positions of the cross-sectional shape, FIG. 7 (b the scanning direction of the approximate curve shown in) (first approximation curve) to calculate the Z _1. Further, as shown in FIG. 8 (a), parallel to the upper end to the slit direction of the correction region X of the chuck after the main surface shape of the substrate for a mask blank, the central, from the height information in each linear Y ₂ of the lower end of the slit direction By calculating the cross-sectional shape of the substrate and calculating a quadratic curve with respect to these three cross-sectional shapes by the least square method, an approximate curve (second approximate curve) Z _{2 in the} slit direction shown in FIG. Is calculated. Then, an approximate curved surface Z ₃ as shown in FIG. 9A is calculated from the approximate curve (first approximate curve) Z ₁ in the scan direction and the approximate curve (second approximate curve) Z _{2 in the} slit direction, and the simulation is performed. Subtract from the post-chuck shape obtained in step 1 above. The substrate shape after subtracting the approximated surface Z ₃ shown in Figure 9 (b).

  Next, the calculated flatness in the correction region (132 mm × 132 mm) of the corrected main surface was calculated, and a substrate having a second threshold value (0.16 μm) or less was selected (ST19). The second threshold value is a standard flatness for which the mask blank substrate is required. As a result, the number of glass substrates satisfying this condition was 97 out of 98. When a substrate satisfying the condition of 0.16 μm or less, which is the same as the second threshold, is selected for the flatness obtained from the post chuck main surface shape (ST15) calculated in the conventional simulation (ST14), 90 of 98 substrates are selected. Therefore, it can be seen that the production yield is greatly improved by performing the main surface shape correction (ST18).

Next, on the glass substrate obtained as described above, as a thin film (light shielding film) for forming a transfer pattern, a MoSiON film (back surface antireflection layer), MoSi (light shielding layer), MoSiON film (antireflection layer) ) To form a mask blank (ST20). Specifically, using a target of Mo: Si = 21: 79 (atomic% ratio), Ar, O ₂ , N ₂ and He were sputtered at a gas pressure of 0.2 Pa (gas flow ratio Ar: O ₂ : N ₂ : He = 5: 4: 49: 42), the power of the DC power source is 3.0 kW, a film made of molybdenum, silicon, oxygen, and nitrogen (MoSiON film) is formed to a thickness of 7 nm, and then the same target is formed. Using Ar and He at a sputtering gas pressure of 0.3 Pa (gas flow ratio Ar: He20: 120), and a DC power supply of 2.0 kW, a film made of molybdenum and silicon (MoSi film: Mo and Si in the film) Is formed with a film thickness of 30 nm, and then a target of Mo: Si = 4: 96 (atomic% ratio) is used, and Ar, O ₂ , N _2, and He are sputtered. No pressure 1 Pa (gas flow ratio _{Ar: O 2: N 2:} He = 6: 5: 11: 16) and then, at 3.0kW power of the DC power supply, molybdenum, silicon, oxygen, film made of nitrogen (MoSiON film) The film was formed with a thickness of 15 nm. The total film thickness of the thin film (light-shielding film) was 52 nm. The back surface antireflection layer, the light shielding layer, and the surface antireflection layer formed under these conditions had very low stress throughout the light shielding film, and the change in the shape of the substrate could be minimized.

  Further, the photomask (binary mask) was produced by patterning the light shielding film and the antireflection film of the mask blank into a predetermined pattern (ST21). The obtained photomask was verified with an exposure apparatus capable of correcting the main surface shape in the scanning direction and the slit direction. After chucking on the mask stage of the exposure apparatus and correcting the main surface shape in the scanning direction and the slit direction, the photomask pattern was transferred onto the resist film of the semiconductor wafer W by exposure. As a result of verifying the CD accuracy and pattern position accuracy of the pattern transferred to the resist film, it was confirmed that this photomask can sufficiently cope with the DRAM hp32 nm generation.

Example 4
As in Example 1, the steps from ST11 to ST17 were performed, 98 glass substrates were selected, and main surface shape correction (ST18) in the scanning direction (first direction) was performed. Next, the calculated flatness in the correction region (132 mm × 132 mm) of the corrected main surface was calculated, and a substrate having a second threshold value (0.08 μm) or less was selected (ST19). The second threshold value is a standard flatness for which the mask blank substrate is required. As a result, the number of glass substrates satisfying this condition was 92 out of 98. When a substrate satisfying the condition of 0.08 μm or less, which is the same as the second threshold, is selected for the flatness obtained from the post chuck main surface shape (ST15) calculated in the conventional simulation (ST14), 84 of 98 substrates are selected. Therefore, it can be seen that the production yield is greatly improved by performing the main surface shape correction (ST18).

  Next, on the glass substrate obtained as described above, as in Example 1, as a thin film for forming a transfer pattern, a light-shielding film composed of a back-surface antireflection layer, a light-shielding layer, and a surface antireflection layer is formed. Then, a mask blank was manufactured (ST20). Further, a photomask (binary mask) was manufactured by patterning the light shielding film of the mask blank into a predetermined pattern (ST21). The obtained photomask was verified with an exposure apparatus capable of at least correcting the main surface shape in the scanning direction. After chucking on the mask stage of the exposure apparatus and correcting the main surface shape in the scanning direction, the resist film on the semiconductor wafer W was exposed to transfer a photomask pattern. As a result of verifying the CD accuracy and pattern position accuracy of the pattern transferred to the resist film, it was confirmed that this photomask can sufficiently cope with the DRAM hp22 nm generation.

(Example 5)
Similarly to Example 2, the steps from ST11 to ST17 were performed, 98 glass substrates were selected, and main surface shape correction (ST18) in the slit direction (second direction) was performed. Next, the calculated flatness in the correction region (132 mm × 132 mm) of the corrected main surface was calculated, and a substrate having a second threshold value (0.08 μm) or less was selected (ST19). The second threshold value is a standard flatness for which the mask blank substrate is required. As a result, the number of glass substrates satisfying this condition was 91 out of 98. When a substrate satisfying the condition of 0.08 μm or less, which is the same as the second threshold, is selected for the flatness obtained from the post chuck main surface shape (ST15) calculated in the conventional simulation (ST14), 84 of 98 substrates are selected. Therefore, it can be seen that the production yield is greatly improved by performing the main surface shape correction (ST18).

  Next, on the glass substrate obtained as described above, as in Example 2, as a thin film for forming a transfer pattern, from a phase shift film, a back surface antireflection layer, a light shielding layer, and a surface antireflection layer A light shielding film was formed to manufacture a mask blank (ST20). Further, the light shielding film and the phase shift film of the mask blank were patterned into a predetermined pattern to produce a photomask (phase shift mask) (ST21). The obtained photomask was verified with an exposure apparatus capable of at least correcting the main surface shape in the slit direction. After chucking on the mask stage of the exposure apparatus and correcting the main surface shape in the slit direction, exposure was performed on the resist film of the semiconductor wafer W to transfer the photomask pattern. As a result of verifying the CD accuracy and pattern position accuracy of the pattern transferred to the resist film, it was confirmed that this photomask can sufficiently cope with the DRAM hp22 nm generation.

(Example 6)
As in Example 3, the steps from ST11 to ST17 are performed, 98 glass substrates are selected, and main surface shape correction (ST18) in the scanning direction (first direction) and slit direction (second direction) is performed. went. Next, the calculated flatness in the correction region (132 mm × 132 mm) of the corrected main surface was calculated, and a substrate having a second threshold value (0.08 μm) or less was selected (ST19). The second threshold value is a standard flatness for which the mask blank substrate is required. As a result, the number of glass substrates satisfying this condition was 93 out of 98. When a substrate satisfying the condition of 0.08 μm or less, which is the same as the second threshold, is selected for the flatness obtained from the post chuck main surface shape (ST15) calculated in the conventional simulation (ST14), 84 of 98 substrates are selected. Therefore, it can be seen that the production yield is greatly improved by performing the main surface shape correction (ST18).

  Next, on the glass substrate obtained as described above, as in Example 3, a light-shielding film comprising a back-surface antireflection layer, a light-shielding layer, and a surface antireflection layer is formed as a thin film for forming a transfer pattern. Then, a mask blank was manufactured (ST20). Further, a photomask (binary mask) was manufactured by patterning the light shielding film of the mask blank into a predetermined pattern (ST21). The obtained photomask was verified with an exposure apparatus capable of correcting the main surface shape in both the scanning direction and the slit direction. After chucking on the mask stage of the exposure apparatus and correcting the main surface shape in the scanning direction and the slit direction, the photomask pattern was transferred onto the resist film of the semiconductor wafer W by exposure. As a result of verifying the CD accuracy and pattern position accuracy of the pattern transferred to the resist film, it was confirmed that this photomask can sufficiently cope with the DRAM hp22 nm generation.

  The present invention is not limited to the above embodiment, and can be implemented with appropriate modifications. For example, the material, size, processing procedure, and the like in the above embodiment are merely examples, and various modifications can be made within the scope of the effects of the present invention. In addition, various modifications can be made without departing from the scope of the object of the present invention.

DESCRIPTION OF SYMBOLS 1 Mask stage 1a Chuck 2 Photomask 3 Slit member 3a Slit 4 Light source 5 Illumination optical system 6 Reduction optical system 7 Wafer stage W Semiconductor wafer

Claims (16)

A method of manufacturing a mask blank substrate used for a photomask chucked on a mask stage of an exposure apparatus,
Measuring the pre-chuck main surface shape in the measured region on the main surface of the substrate;
Obtaining a post-chuck main surface shape of the substrate by simulation based on the pre-chuck main surface shape of the substrate and the shape of the mask stage when the photomask produced from the substrate is set in the exposure apparatus;
For the substrate, a first approximate curve that approximates a cross-sectional shape along a first direction within a correction region of the post-chuck main surface shape is calculated, an approximate curved surface is calculated from the first approximate curve, and the post-chuck main surface Performing a correction to be subtracted from the shape and calculating a corrected main surface shape;
Selecting a flatness within a correction region of the corrected main surface shape that is a predetermined threshold value or less;
A method for manufacturing a mask blank substrate, comprising:   The step of calculating the corrected main surface shape calculates a second approximate curve that approximates a cross-sectional shape along a second direction orthogonal to the first direction, calculates an approximate curved surface from the second approximate curve, 2. The correction for subtracting the approximate curved surface calculated from the second approximated surface from the post-chuck main surface shape after performing the correction for subtracting the approximated curved surface calculated from the first approximated curve is performed. The manufacturing method of the board | substrate for mask blanks as described. A method of manufacturing a mask blank substrate used for a photomask chucked on a mask stage of an exposure apparatus,
Measuring the pre-chuck main surface shape in the measured region on the main surface of the substrate;
Obtaining a post-chuck main surface shape of the substrate by simulation based on the pre-chuck main surface shape of the substrate and the shape of the mask stage when the photomask produced from the substrate is set in the exposure apparatus;
For the substrate, a first approximate curve that approximates the cross-sectional shape along the first direction within the correction region of the post-chuck main surface shape is calculated, and approximated to the cross-sectional shape along the second direction orthogonal to the first direction Calculating a second approximated curve, calculating an approximated curved surface from the first approximated curve and the second approximated curve, and subtracting it from the post-chuck main surface shape, and calculating a corrected main surface shape;
Selecting a flatness within a correction region of the corrected main surface shape that is a predetermined threshold value or less;
A method for manufacturing a mask blank substrate, comprising:   4. The method of manufacturing a mask blank substrate according to claim 1, wherein the first approximate curve is a quadratic curve or a quartic curve.   5. The method for manufacturing a mask blank substrate according to claim 2, wherein the second approximate curve is a quadratic curve or a quartic curve.   6. The method for manufacturing a mask blank substrate according to claim 1, wherein the predetermined threshold value is 0.16 [mu] m.   The method for manufacturing a mask blank substrate according to claim 1, wherein the predetermined threshold is 0.08 μm.   The method for manufacturing a mask blank substrate according to claim 1, wherein the actual measurement region is a region including the correction region.   The method for manufacturing a mask blank substrate according to claim 1, wherein the correction region is a region within a 132 mm square.   10. The mask blank according to claim 1, further comprising a step of selecting a substrate having a flatness of 0.4 μm or less in the actual calculation region of the pre-chuck main surface shape. 11. A method for manufacturing a substrate.   The method for manufacturing a mask blank substrate according to claim 10, wherein the actual calculation region is a region including the correction region.   The method for manufacturing a mask blank substrate according to claim 10, wherein the actual calculation area is an area within a 142 mm square.   13. The exposure apparatus according to claim 1, wherein exposure light is irradiated to the photomask through a slit that is movable in the first direction and extends in the second direction. A method for producing a mask blank substrate according to any one of the above.   A method for manufacturing a mask blank, comprising: forming a thin film on a main surface on the side of the main surface shape before chucking of the mask blank substrate obtained by the method according to any one of claims 1 to 13. .   A method for producing a photomask, comprising forming a transfer pattern on a thin film of a mask blank obtained by the method according to claim 14.   A step of chucking the photomask obtained by the method according to claim 15 to a mask stage of an exposure apparatus capable of performing main surface shape correction, and exposing and transferring a photomask pattern onto a resist film on a wafer. A method of manufacturing a semiconductor device.

Images