Method, computer program product and control device for determining a repair method for a defect in a photomask for microlithography and method for particle beam-induced processing of a defect in a photomask for microlithography

Abstract

A method for determining a repair shape (406) of a defect (D) of a photomask (100) for microlithography, comprising the steps: a) Providing (S1) an image (400) or at least part of the photomask (100), b) Determine (S2) at least one section of a boundary line (408) of the defect (D) based on the image (400), c) Determining (S3) a curvature (K1) of at least one section of the boundary line (408) at at least one position (P1) of the at least one section of the boundary line (408), d) Determine (S4) at least one correction value (K1) for at least one position (P1) based on the determined curvature (K1), and e) Determine (S5) at least one corrected section of the boundary line (408) by correcting the at least one section of the boundary line (408) based on the at least one determined correction value (K1).

Description

[0001] The present invention relates to a method, a computer program, and a control device for determining a repair method for a defect in a photomask for microlithography. Furthermore, the present invention relates to a method for particle beam-induced processing of a defect in a photomask for microlithography.

[0002] Microlithography is used to manufacture microstructured components, such as integrated circuits. The microlithography process is carried out using a lithography system, which includes an illumination system and a projection system. The image of a mask (reticule) illuminated by the illumination system is projected by the projection system onto a substrate, such as a silicon wafer, coated with a photosensitive layer (photoresist) and positioned in the image plane of the projection system. This transfers the mask structure onto the photosensitive coating of the substrate.

[0003] Driven by the pursuit of ever smaller structures in the production of integrated circuits, EUV lithography systems are currently being developed which use light with a wavelength in the range of 0.1 nm to 30 nm, in particular 13.5 nm.

[0004] The photomasks (i.e., lithography masks) used in microlithography feature microstructures that are projected onto the substrate on a reduced scale. The photomask structures themselves are already very small, ranging in size from a few micrometers to a few nanometers. To produce microstructured devices with high accuracy using microlithography, the photomasks must also be manufactured with high precision and be free of defects (e.g., faulty structures and impurities). A defect-free photomask is also crucial for microlithography because a single photomask is typically used for numerous exposures. Therefore, photomasks undergo extensive inspection for defects, and any defects found are repaired.

[0005] Typical defects in photomasks include the absence of intended structures and the presence of unintended structures. Such defects can be caused, for example, by an unsuccessful etching process, such as one that progressed too quickly or acted on the wrong location. These defects can be eliminated by selectively etching excess material or selectively depositing (depositing) additional material at the relevant locations. This can be achieved very precisely using particle beam-induced processing (e.g., focused electron beam induced processing, FEBIP), as described, for example, in DE 10 2017 208 114 A1. In this process, a particle beam and a process gas (e.g., etching gas or depositing gas) are applied to the area of ​​the photomask to be processed.The particle beam activates a local chemical reaction between a material of the photomask and the process gas, causing material to be locally removed from or deposited onto the photomask.

[0006] In particle beam-induced processing of a photomask, the precise determination of a repair shape for the defect is crucial for the processing process. From US 11,079,674 B2, it is known to correct a determined pre-repair shape for a photomask defect by taking into account the position of a desired photomask structure relative to the photomask defect.

[0007] Furthermore, DE 10 2021 123 440 A1 discloses a method for particle beam-induced processing of a defect in a photomask for microlithography, comprising the steps: a1) providing an image of at least a part of the photomask, b1) determining a geometric shape of a defect in the image as a repair shape, c1) subdividing the repair shape into a number of n pixels according to a first rasterization, d1) subdividing the repair shape into a number of m pixels according to a second rasterization, wherein the second rasterization results from a subpixel shift of the first rasterization, e1) providing an activating particle beam and a process gas at each of the n pixels of the repair shape according to the first rasterization, and f1) providing the activating particle beam and the process gas at each of the m pixels of the repair shape according to the second rasterization.

[0008] Against this background, one object of the present invention is to better determine a repair method for a defect in a photomask for microlithography.

[0009] According to a first aspect, a method for determining a repair method for a defect in a photomask for microlithography is proposed. The method comprises the following steps: a) Providing an image of at least part of the photomask, b) Determining at least one section of a boundary line of the defect based on the image, c) Determining a curvature of at least one section of the boundary line at at least one position of the boundary line, d) Determine at least one correction value for at least one position based on the determined curvature, and e) Determining at least one corrected section of the boundary line by correcting at least one section of the boundary line based on the at least one determined correction value.

[0010] By determining the repair shape based on the curvature of the boundary line of at least one segment of the boundary line (e.g., a boundary line of a pre-repair shape of the defect), an improved repair shape for a photomask defect can be determined. This allows for better processing and repair of the photomask defect. For example, an unwanted structure in the defect area can be better (e.g., completely) removed, and / or damage to adjacent areas (e.g., an adjacent capping layer) can be avoided and / or reduced. Alternatively, a missing structure in the defect area can be better (e.g., completely) restored.

[0011] Insofar as the following refers to determining a pre-repair shape of the defect, a boundary line of the pre-repair shape, a curvature of the boundary line of the pre-repair shape, and determining the repair shape by correcting the pre-repair shape, this can also refer to determining at least a section of a boundary line of the defect (i.e., without determining the entire pre-repair shape), at least a section of a boundary line of the defect, a curvature of at least one section of the boundary line, and determining the repair shape by correcting at least one section of the boundary line.

[0012] To repair the photomask defect, the repair shape is divided into, for example, a number of m pixels. A particle beam is then positioned at each pixel of the repair shape (e.g., for a predetermined dwell time). This particle beam is directed, for example, towards the center of a pixel. The particle beam has a Gaussian intensity distribution, the intensity of which decreases to a predetermined level by the pixel's edge. Due to this Gaussian intensity distribution, the particle beam intensity is also greater than zero for pixels adjacent to the repaired pixel. In other words, when the particle beam is positioned at a specific pixel, a residual particle beam dose also affects neighboring pixels.

[0013] At the edge of the pre-repair shape, pixels of the pre-repair shape that are to be processed are located adjacent to pixels outside the pre-repair shape that are not to be processed. In a concave edge region of the pre-repair shape, pixels outside the pre-repair shape that are not to be processed are surrounded by a particularly large number of pixels at the edge of the pre-repair shape that are to be processed. Therefore, in a concave edge region of the pre-repair shape, an excessively high particle beam dose can result for the adjacent pixels not to be processed, thus increasing the risk of damage to adjacent areas (e.g., an adjacent cover layer).

[0014] Similarly, in a convex border region of the pre-repair shape, there are only a few pixels to be processed, so the particle beam dose in the area of ​​the convex shape of the pre-repair shape may be insufficient for complete processing of the defect at that location. In particular, in the case of etching, it may happen that material to be removed is not removed (i.e., so-called residues remain).

[0015] By determining at least one correction value based on the curvature of the edge line of the pre-repair shape and correcting the pre-repair shape based on the at least one determined correction value to determine the repair shape, the described unfavorable particle beam dose in the edge region of the repair shape can be counteracted.

[0016] The photomask for microlithography is, for example, a photomask for an EUV lithography system. EUV stands for "extreme ultraviolet" and refers to a wavelength of the working light between 0.1 nm and 30 nm, specifically 13.5 nm. In an EUV lithography system, EUV radiation is directed onto a photomask (reticle) by means of a beam shaping and illumination system. This photomask is designed as a reflective optical element (reflective photomask). The photomask has a structure that is reduced in size and projected onto a wafer or similar substrate by means of a projection system within the EUV lithography system.

[0017] The photomask for microlithography can, for example, also be a photomask for a DUV lithography system. DUV stands for "deep ultraviolet" and refers to a wavelength of the working light between 30 nm and 250 nm, particularly 193 nm or 248 nm. In a DUV lithography system, DUV radiation is directed onto a photomask, which is specifically designed as a transmissive optical element (transmissive photomask), by means of a beam shaping and illumination system.

[0018] A photomask for microlithography, for example, comprises a substrate and a structure formed on the substrate by a coating. The photomask can be, for instance, a reflective photomask for use in EUV lithography, where the pattern to be imaged is realized as an absorbing (i.e., opaque or partially opaque) coating on a reflective substrate. Alternatively, the photomask, particularly for use in DUV lithography, can also be, for example, a transmissive photomask, where the pattern to be imaged is realized as an absorbing coating on a transparent substrate. The photomask can also be a mask for nanoimprint lithography (NIL).

[0019] For example, in the case of a photomask for an EUV lithography system, the substrate comprises an alternating sequence of molybdenum and silicon layers. In the case of a transmissive photomask for DUV lithography, the substrate can also include, for example, silicon dioxide (SiO2), such as fused silica. The structured coating may include, for example, chromium, chromium compounds, tantalum compounds, and / or compounds of silicon, nitrogen, oxygen, and / or molybdenum. The substrate and / or the coating may also include other materials.

[0020] The proposed method allows for the detection and localization of a photomask defect, particularly a defect in a structured coating of the photomask, within an image. A defect is, in particular, an incorrectly applied (e.g., absorbing or reflecting) coating of the photomask to the substrate.

[0021] To determine the repair method for the defect, an image of at least part of the photomask is provided in step a).

[0022] The image of at least part of the photomask is acquired, for example, using a scanning electron microscope (SEM). This image of at least part of the photomask has a spatial resolution in the range of a few nanometers. The image can also be acquired with a scanning probe microscope (SPM), such as an atomic force microscope (AFM) or a scanning tunneling microscope (STM).

[0023] The method may in particular include a step of capturing the image of at least part of the photomask, e.g. by means of a scanning electron microscope and / or a scanning probe microscope.

[0024] In step b), a geometric shape (e.g., a two-dimensional geometric shape) of the defect is determined based on the provided image of at least part of the photomask. For example, the geometric shape is determined in the provided image or in an image (e.g., a difference image) determined based on the provided image. The geometric shape of the defect determined in step b) is referred to herein as the pre-repair shape. The pre-repair shape corrected by the proposed method is referred to herein as the repair shape (final repair shape).

[0025] Determining the geometric shape of a defect based on the image in step b) can involve subtracting the image, or a portion thereof, from a reference image to obtain a difference image. Specifically, the image, or portion thereof, and the reference image exhibit the same structures, except for the defect. That is, a geometric shape of structures in the image, or portion thereof, corresponds to a geometric shape of structures in the reference image. For example, the photomask exhibits recurring structures (e.g., absorber structures), and a portion of the provided image containing the defect is compared to another portion of the provided image containing the same structural pattern but no defect. In this case, the other portion of the provided image serves as the reference image.In other examples, however, the reference image can be provided in a different way. By subtracting the reference image from the image or the portion of the image containing the defect, a difference image is obtained in which only the defect is depicted. The geometric shape of the defect is then determined, for example, from this difference image.

[0026] Determining the geometric shape of the defect includes, for example, edge detection in the provided image and / or the difference image.

[0027] An identified and / or detected edge of the defect forms, in particular, the boundary line of the pre-repair mold. The boundary line is, in particular, a circumferential (e.g., closed) boundary line. An area enclosed by the boundary line is, in particular, an area of ​​the pre-repair mold. The boundary line is, in particular, a limit line of the pre-repair mold.

[0028] The determined edge line of the pre-repair shape can, for example, be smoothed in step b). Then step c) can be carried out, especially for the smoothed edge line.

[0029] In step c), the curvature of the boundary line is determined at at least one position along the boundary line. The curvature of the boundary line can also be determined at multiple positions along the boundary line and / or for the entire boundary line.

[0030] In this context, curvature refers to the mathematical curvature of the boundary line. Specifically, curvature means a deviation of the boundary line from a straight line. Curvature can have a (mathematical) magnitude (i.e., a curvature measure) and a direction (positive or negative). Here, the direction of curvature is considered to be counterclockwise. The greater the magnitude of the curvature, the stronger the curvature.

[0031] In the case of a negative curvature of the boundary line (counterclockwise to the right), the boundary line exhibits a negative second derivative. Furthermore, the pre-repair shape has a concave section in the area of ​​the negative curvature of the boundary line.

[0032] When the boundary line has a positive curvature (counterclockwise to the left), the boundary line exhibits a positive second derivative. Furthermore, the pre-repair shape has a convex section in the region of the positive curvature of the boundary line.

[0033] For example, in step c), a (mathematical) magnitude of the curvature of the boundary line is determined at at least one position. Furthermore, in step c), it is determined, for example, whether the determined magnitude of the curvature at at least one position is greater than zero (i.e., curvature is present) and / or greater than a predetermined threshold. In this case, steps d) and e) are executed, for example, only if the determined magnitude of the curvature is greater than zero and / or greater than the predetermined threshold.

[0034] Determining the curvature of the boundary line at at least one position in step c) is based, for example, on known image analysis methods and / or algorithms for contour detection, edge detection, and curvature determination. For example, determining the curvature of the boundary line is based on known methods in programming language libraries and computer programs.

[0035] For the analytical or numerical determination of contours, a method can be applied that extracts one or more contours from a pre-repair mold, such that the extracted contours are in the form x(t) and y(t). The functions x(t) and y(t) can then be filtered, for example, with a second-order Savitzky-Golay filter. This yields the derivatives x'(t), x''(t) and y'(t), y''(t). Curvature can then be calculated, for example, from the well-known formula for a plane curve: κ(t)=x˙(t)y¨(t)−x¨(t)y˙(t)(x˙(t)2+y˙(t)2)32.

[0036] In step d), the at least one correction value for the at least one position of the boundary line is determined based on the calculated curvature of the boundary line at that at least one position. For example, the at least one correction value for the at least one position of the boundary line specifies how (e.g., to what extent and in which direction) the boundary line of the pre-repair shape should be shifted at that at least one position, depending on the curvature present there, in order to determine the repair shape (final repair shape). For example, the correction value is a real number whose absolute value indicates the extent of the shift and whose sign indicates the direction of the shift.

[0037] In step e), the pre-repair shape is corrected to determine the repair shape (final repair shape). In particular, in step e), a new boundary line is determined that surrounds the repair shape (final repair shape).

[0038] Steps a) to e) are carried out using computer-implemented methods.

[0039] According to one embodiment: In step b), a geometric shape of the defect is determined based on the image as a pre-repair shape with a border line that has at least one section of the border line. In step c), a curvature of the boundary line is determined at at least one position of the boundary line. In step d), at least one correction value is determined for at least one position based on the determined curvature, and In step e), the repair shape is determined by correcting the pre-repair shape based on the at least one determined correction value.

[0040] According to another embodiment, the at least one correction value specifies a displacement of the boundary line at the at least one position in a direction substantially perpendicular to a tangent to the boundary line at the at least one position.

[0041] The direction essentially perpendicular to the tangent to the boundary line at the at least one position includes, for example, a direction that deviates by ±30° or less, ±20° or less, ±10° or less and / or ±5° or less from a direction exactly perpendicular to the tangent.

[0042] If a curvature of zero is determined at a position on the boundary line, this means that no curvature of the boundary line is determined at that position. Then, for example, step d) (and step e) is not performed, or the correction value determined in step d) is zero, meaning that the displacement of the boundary line is also zero.

[0043] If a curvature is detected at a position on the boundary line whose (mathematical) magnitude is less than a predetermined threshold, then this means, for example, that no significant curvature of the boundary line is detected at that position. In that case, step d) (and step e) is not performed.

[0044] According to a further embodiment, in step c) the magnitude and sign of the curvature of the boundary line are determined at the at least one position of the boundary line. Furthermore, in step d) the at least one correction value for the at least one position of the boundary line is determined based on the determined magnitude and sign of the curvature.

[0045] This allows the pre-repair shape to be corrected more strongly at at least one position of the boundary line the greater the magnitude of the curvature of the boundary line at that position. Furthermore, the sign of the curvature is taken into account during the correction.

[0046] The magnitude of the curvature is, in particular, a mathematical magnitude, i.e., an absolute value, of the boundary line. The sign of the curvature is determined by the value of the second derivative of the boundary line at at least one position. If the second derivative of the boundary line at at least one position has a negative value, then the curvature has a negative sign and is concave up. If the second derivative of the boundary line at at least one position has a positive value, then the curvature has a positive sign and is concave down.

[0047] According to a further embodiment, in step c) a second derivative of the boundary line of the pre-repair mold is determined. Furthermore, in step c) the magnitude of the curvature of the boundary line at the at least one position is determined based on the second derivative.

[0048] In this context, a derivative (e.g., second derivative) is understood to be a mathematical derivative with respect to position.

[0049] The second derivative of the boundary line gives, in particular, a measure of the curvature of the boundary line.

[0050] In particular, the magnitude of the curvature of the boundary line at at least one position is determined based on a value of the second derivative at at least one position.

[0051] According to a further embodiment, in step c) a second derivative of the boundary line of the pre-repair mold is determined. Furthermore, in step c) a sign of the curvature of the boundary line at the at least one position is determined based on the second derivative.

[0052] The second derivative of the boundary line indicates, in particular, a sign (positive or negative) of the curvature of the boundary line.

[0053] In particular, the sign of the curvature of the boundary line is determined at at least one position based on a value of the second derivative at at least one position.

[0054] If the second derivative of the boundary line is negative, i.e. less than zero, at at least one position of the boundary line, then there is a negative curvature at at least one position.

[0055] If the second derivative of the boundary line is positive, i.e. greater than zero, at at least one position of the boundary line, then there is a positive curvature of the boundary line at at least one position.

[0056] According to another embodiment, the repair shape is determined in step e) such that: For a position of the boundary line of the pre-repair shape for which a positive curvature is determined, the boundary line at this position is shifted so that the pre-repair shape is enlarged at this position, and For a position of the boundary line of the pre-repair shape for which a negative curvature is determined, the boundary line at this position is shifted so that the pre-repair shape is reduced in size at this position.

[0057] A positive curvature of the edge of the pre-repair shape at at least one position means that the pre-repair shape has a convex section in this area. A convex section of the pre-repair shape is, in particular, a section that projects outwards with respect to the pre-repair shape. With an outwardly projecting section of the pre-repair shape, it is advantageous to enlarge the pre-repair shape at this point in order to achieve a sufficient particle beam dose during processing with a particle beam.

[0058] It can also be said that for a position of the boundary line of the pre-repair shape for which a positive curvature is determined, the boundary line at this position is shifted outwards with respect to the pre-repair shape.

[0059] In particular, for a position of the boundary line for which a positive curvature of the boundary line is determined, a correction value is determined in step d) which indicates a shift of the boundary line at this position, which, when correcting the pre-repair shape in step e), leads to an enlargement of the repair shape.

[0060] A negative curvature of the edge of the pre-repair mold at at least one position means that the pre-repair mold has a concave section in this area. A concave section of the pre-repair mold is, in particular, a section that is hollowed out inwards with respect to the pre-repair mold and / or has a recess. With a concave section of the pre-repair mold, it is advantageous to reduce the size of the pre-repair mold at this point to prevent an excessively high particle beam dose during processing with a particle beam at this location.

[0061] It can also be said that for a position of the boundary line of the pre-repair shape for which a negative curvature of the boundary line is determined, the boundary line at this position is shifted inwards with respect to the pre-repair shape.

[0062] In particular, for a position of the boundary line for which a negative curvature of the boundary line is determined, a correction value is determined in step d) which indicates a shift of the boundary line at this position, which, when correcting the pre-repair shape in step e), leads to a reduction in the size of the repair shape.

[0063] According to a further embodiment, in step c) the magnitude of the curvature of the boundary line at at least one position is determined. Furthermore, in step c) it is determined whether the magnitude of the curvature is greater than zero and / or greater than a predetermined threshold. In addition, in step e) the repair shape is determined such that, for a position of the boundary line of the pre-repair shape for which the determined magnitude of the curvature is greater than zero and / or greater than the predetermined threshold, the boundary line at that position is shifted by an amount that is a mathematical function of the magnitude of the curvature at that position.

[0064] For example, the boundary line at the position in question is shifted by an amount that is a linear function of the magnitude of the determined curvature at that position and / or that is proportional to a power of the magnitude of the determined curvature (i.e., a monomial with the magnitude of the determined curvature as a variable) at that position.

[0065] For example, the at least one correction value for the at least one position of the boundary line can be determined based on the following equation: X=A⋅B⋅Cn.

[0066] In this equation, X denotes the correction value for a position of the boundary line, B a sign of the curvature at that position of the boundary line (i.e., B can take the values ​​+1 or -1), and C a magnitude of the curvature at that position of the boundary line.

[0067] Furthermore, in the equation above, the parameter A is a scaling factor and the parameter n is a scaling exponent. The scaling factor A and the scaling exponent n are each, in particular, positive real numbers. For n=1, the equation simplifies to X = A · B · C, and the correction value X is directly proportional (proportionality factor A) to the curvature (B · C), i.e., a linear function of the curvature (B · C). For n≠1, the equation is X = A · B · C n and the correction value X is directly proportional (proportionality factor A) to a power n of the magnitude C of the curvature.

[0068] According to a further embodiment, the at least one correction value in step d) is determined based on one or more pre-determined scaling parameters, which indicate a dependence of an extent of a displacement of the boundary line at the at least one position on an amount of the determined curvature at the at least one position.

[0069] The one or more pre-determined scaling parameters include, for example, the scaling factor A described above and / or the scaling exponent n described above.

[0070] According to another embodiment, one or more predetermined scaling parameters have a predetermined scaling factor and / or a predetermined scaling exponent.

[0071] The pre-determined scaling factor indicates a linear dependence of the extent of the displacement of the boundary line at the at least one position on the amount of the determined curvature at the at least one position.

[0072] The pre-determined scaling exponent indicates a dependence of the extent of the displacement of the boundary line at the at least one position on the magnitude of the determined curvature at the at least one position such that the extent of the displacement is proportional to a power of the determined magnitude of the curvature.

[0073] The one or more pre-determined scaling parameters can, for example, depend on parameters of the photomask, e.g., a material of the photomask, and parameters of the processing, e.g., an accelerating voltage of the particle beam column, etc., in addition to the curvature of the boundary line.

[0074] According to another embodiment: The determined pre-repair shape is divided into a number of k pixels, The pixels of the pre-repair shape that are touched and / or intersected by the boundary line form the boundary pixels of the pre-repair shape. The at least one position of the border line corresponds to the position of at least one border pixel, and In step e), the repair shape is corrected by adding or removing one or more pixels adjacent to or from the pre-repair shape, based on the at least one determined correction value.

[0075] For particle beam-induced processing of the repair shape, k pixels are defined in the pre-repair shape and / or m pixels are defined in the repair shape. The subdivision of the pre-repair shape into k pixels and the subdivision of the repair shape into m pixels is based, for example, on the same rasterization. For instance, the provided image or a portion thereof, which is larger than the pre-repair shape, is subdivided into l pixels using a predetermined rasterization. Here, the k pixels of the pre-repair shape and the m pixels of the repair shape each form a subset of the l pixels of the provided image or image portion. That is, the number l is larger than both the number k and the number m.

[0076] If the pre-repair shape is reduced in size overall during the correction process in step d), then the repair shape is smaller than the pre-repair shape, and the number m of pixels in the repair shape is less than the number k of pixels in the pre-repair shape. Conversely, if the pre-repair shape is increased in size overall during the correction process in step d), then the repair shape is larger than the pre-repair shape, and the number m of pixels in the repair shape is greater than the number k of pixels in the pre-repair shape.

[0077] During particle beam-induced processing of the repair shape, the particle beam is directed onto each of the m pixels of the repair shape (e.g., for a predetermined dwell time). Specifically, an intensity maximum of the particle beam is directed onto the center point of each of the m pixels of the repair shape. In other words, the m pixels of the repair shape represent a raster, particularly a two-dimensional raster, of the repair shape for particle beam-induced processing. The m pixels of the repair shape correspond, for example, to the impact surfaces of the particle beam during particle beam-induced processing of the defect. For example, a pixel size of the m pixels (e.g., also of the k pixels and l pixels mentioned above) is chosen such that an intensity distribution of a particle beam directed onto the center of a pixel is determined by its (e.g.,The intensity distribution (Gaussian) has decayed to a predetermined intensity value by the edge of the pixel. The predetermined intensity value can, for example, correspond to a fall (i.e., a decrease) to half the intensity maximum or to another fraction of the particle beam's intensity maximum. A pixel size and / or a particle beam half-width is, for example, sub-nanometers or a few nanometers.

[0078] The pixels of the pre-repair shape that lie on the boundary line (i.e., are touched and / or intersected by the boundary line) form boundary pixels of the pre-repair shape.

[0079] The at least one position of the boundary line corresponds to the position of at least one boundary pixel. The position of the at least one boundary pixel, for example, has a center / midpoint. The position of the at least one boundary pixel can also have the midpoint of a segment of the boundary line that is covered by the at least one boundary pixel.

[0080] In step e), one or more pixels are added to or removed from the pre-repair shape. The added or removed pixels are located adjacent to (e.g., near) the at least one edge pixel for which the curvature was determined.

[0081] According to a further embodiment, in step e), for an edge pixel for which negative curvature is determined, this edge pixel and / or further adjacent pixels are removed from the pre-repair shape in a direction substantially perpendicular to a tangent to the edge line at this edge pixel. Furthermore, for an edge pixel for which positive curvature is determined, one or more further pixels are added to the pre-repair shape in a direction substantially perpendicular to the tangent to the edge line at this edge pixel.

[0082] A second aspect involves proposing a method for particle beam-induced processing of a defect in a photomask for microlithography. The method exhibits: Determining a repair method for the defect according to the procedure described above, in accordance with the first aspect, Dividing the determined repair shape into a number of m pixels, and Providing a particle beam at each of the m pixels of the repair shape to address the defect.

[0083] For example, an activating particle beam and a process gas are provided at each of the m pixels of the repair mold.

[0084] Processing the defect includes, for example, etching the defect, in which material is locally removed from the photomask, or depositing material onto the photomask in the defect area. The proposed method allows, for instance, for more effective etching away of superfluous structures in the defect area, or for more effective restoration of missing structures in the defect area. In particular, the proposed method enables more precise and effective etching of the defect's periphery, prevents unintentional damage to adjacent areas during the etching process, and / or allows for more precise and effective restoration of missing structures in the defect's periphery.

[0085] The proposed method can detect, locate, and repair defects in a photomask, particularly defects in a structured coating. For example, the method can be used to reapply the coating to areas where it is missing. Furthermore, the method can be used to remove the coating from areas of the photomask where it has been incorrectly applied.

[0086] The process gas is, for example, a precursor gas and / or an etching gas. The process gas can, for example, be a mixture of several gaseous components, i.e., a process gas mixture. The process gas can, for example, be a mixture of several gaseous components, each of which contains only one specific type of molecule.

[0087] Suitable precursor gases for the deposition or growth of raised structures include, in particular, alkyl compounds of main group elements, metals, or transition elements. Examples include cyclopentadienyl trimethylplatinum (CpPtMe3 Me = CH4), methylcyclopentadienyl trimethylplatinum (MeCpPtMe3), tetramethyltin (SnMe4), trimethylgallium (GaMe3), ferrocene (Cp2Fe), bis-aryl chromium (Ar2Cr), and / or carbonyl compounds of main group elements, metals, or transition elements, such as chromium hexacarbonyl (Cr(CO)6), molybdenum hexacarbonyl (Mo(CO)6), tungsten hexacarbonyl (W(CO)6), dicobalt octacarbonyl (Co2(CO)8), and triruthenium dodecacarbonyl (Ru3(CO)6). 12), iron pentacarbonyl (Fe(CO)5), and / or alkoxide compounds of main group elements, metals or transition elements, such as tetraethoxysilane (Si(OC2H5)4), tetraisopropoxytitanium (Ti(OC3H7)4), and / or halide compounds of main group elements, metals or transition elements, such as tungsten hexafluoride (WF6), tungsten hexachloride (WCl6), titanium tetrachloride (TiCl4), boron trifluoride (BCl3), silicon tetrachloride (SiCl4), and / or complexes with main group elements, metals or transition elements, such as copper bishexafluoroacetylacetonate (Cu(C5F6HO2)2), dimethyl gold trifluoroacetylacetonate (Me2Au(C5F a H4O2)), and / or organic compounds such as carbon monoxide (CO), carbon dioxide (CO2), aliphatic and / or aromatic hydrocarbons, and the like.

[0088] The etching gas may, for example, comprise: xenon difluoride (XeF₂), xenon dichloride (XeCl₂), xenon tetrachloride (XeCl₄), water vapor (H₂O), heavy water (D₂O), oxygen (O₂), ozone (O₃), ammonia (NH₃), nitrosyl chloride (NOCl), and / or one of the following halide compounds: XNO, XONO₂, X₂O, XO₂, X₂O₄, X₂O₆, where X is a halide. Further etching gases for etching one or more of the deposited test structures are specified in the applicant's US patent 2012 / 0273458A1.

[0089] The process gas may include additional gases, such as oxidizing gases like hydrogen peroxide (H₂O₂), nitrous oxide (N₂O), nitrogen oxide (NO), nitrogen dioxide (NO₂), nitric acid (HNO₃), and other oxygen-containing gases, and / or halides like chlorine (Cl₂), hydrogen chloride (HCl), hydrogen fluoride (HF), iodine (I₂), hydrogen iodide (HI), bromine (Br₂), hydrogen bromide (HBr), phosphorus trichloride (PCl₃), phosphorus pentachloride (PCl₅), phosphorus trifluoride (PF₃), and other halogen-containing gases, and / or reducing gases like hydrogen (H₂), ammonia (NH₃), methane (CH₄), and other hydrogen-containing gases. These additional gases can be used, for example, in etching processes, as buffer gases, as passivating agents, and for similar applications.

[0090] The particle beam is provided, for example, by means of a device for particle beam-induced processing of a photomask for microlithography. The device may comprise: a particle beam source for generating the particle beam; a particle beam guidance device (e.g., a scan unit) configured to direct the particle beam onto a pixel m of the photomask repair shape; a particle beam shaping device (e.g., electron or beam optics) configured to shape, in particular to focus, the particle beam; optionally, at least one storage container configured to store the process gas or at least one gaseous component of the process gas; optionally, at least one gas supply device configured to supply the process gas or at least one gaseous component of the process gas.to provide at least one gaseous component of the process gas with a predetermined gas flow rate at pixel m of the repair mold.

[0091] The particle beam includes, for example, an electron beam and / or an ion beam. In addition to a particle beam, a laser beam can also be provided (e.g., at each pixel of the repair mold).

[0092] For example, an electron beam is provided using a modified scanning electron microscope. For instance, the image of at least part of the photomask is acquired using the same modified scanning electron microscope that provides the electron beam. However, a particle beam other than an electron beam can also be used to process the photomask.

[0093] For example, the particle beam activates a local chemical reaction between a material of the photomask and the process gas, which locally leads to a deposition of material from the gas phase onto the photomask or to a transition of material from the photomask into the gas phase.

[0094] The particle beam is sequentially delivered to each of the m pixels of the repair mold, for example, using the particle beam guidance system. The particle beam remains at each of the m pixels for a predetermined dwell time, for example, 100 ns. After the particle beam has been delivered to all m pixels of the repair mold, this process can be repeated with one or more repetition cycles.

[0095] According to one embodiment of the second aspect, one or more scaling parameters are empirically determined by analyzing and processing a test photomask for microlithography.

[0096] For example, the test photomask has a defect whose outline, appearing in at least one image of the test photomask, has a border with segmentally different curvatures. Alternatively, the test photomask can also have multiple defects with different curvatures. The different curvatures cover, for example, a range of curvatures between a predetermined negative curvature (e.g., (B·C) = -2) and a predetermined positive curvature (e.g., (B·C) = +2) in uniform intervals (e.g., Δ(B·C) = 0.5). However, values ​​other than those mentioned as examples can also be used.

[0097] Then, for each of the several distinct curvatures of the boundary line, several different values ​​of one or more pre-determined scaling parameters are applied when determining the correction values ​​and the repair shape. Subsequently, each of the repair shapes determined in this way is processed using a particle beam (and, for example, process gas). Next, the result of the processing (e.g., the repair) of the corresponding defect and / or the corresponding section of the defect is examined. Then, the value of one or more pre-determined scaling parameters is determined as the optimal value for which the result of the processing of the corresponding defect is optimal (e.g., the highest degree of removal of a superfluous structure with the least etching into a substrate of the photomask, or the highest degree of completion of a missing structure).

[0098] For example, when empirically determining one or more scaling parameters, the result of the processing (e.g., the repair) is also determined for test photomasks made of different materials and / or defects in the test photomask made of different materials (e.g., absorber materials). Furthermore, when empirically determining one or more scaling parameters, one or more parameters of the particle beam processing can also be varied (e.g., the type of process gas used, one or more parameters of the particle beam used, one or more parameters of the rasterization of the pre-repair or repair shape, etc.).

[0099] According to a third aspect, a computer program product is proposed. The computer program product comprises instructions which, when executed by at least one computer, cause it to execute the above-described method for determining a repair method for a defect in a photomask for microlithography and / or the above-described method for particle beam-induced processing of a defect in a photomask for microlithography.

[0100] A computer program product, such as a computer program tool, can be provided or delivered from a server on a network, for example, as a storage medium such as a memory card, USB stick, CD-ROM, DVD, or as a downloadable file. This can be done, for example, in a wireless communication network by transmitting the corresponding file containing the computer program product or tool.

[0101] According to a fourth aspect, a control device for determining a repair method for a defect in a photomask for microlithography is proposed. The control device features: a provisioning unit for providing an image of at least part of the photomask, a first investigative unit to determine at least a section of a boundary line of the defect based on the image, a second investigative unit for determining a curvature of at least one section of the boundary line at at least one position of the at least one section of the boundary line, a third determination unit for determining at least one correction value for at least one position based on the determined curvature, and a fourth investigation unit for determining at least one corrected section of the boundary line by correcting the at least one section of the boundary line based on the at least one determined correction value.

[0102] The control unit is, for example, part of a device for particle beam-induced processing of a photomask for microlithography.

[0103] The control unit is specifically designed to perform the procedure described above in accordance with the first aspect for determining a repair method for a defect in a photomask for microlithography.

[0104] The units mentioned above and below, such as the control unit, the deployment unit, and the investigation units, can each be implemented in hardware and / or software. In the case of a hardware implementation, the corresponding unit can be designed as a device or as part of a device, for example, as a computer or a microprocessor. The device can, for example, include a central processing unit (CPU), a graphics processing unit (GPU), programmable hardware logic (e.g., a field-programmable gate array, FPGA), an application-specific integrated circuit (ASIC), or the like.Furthermore, one or more units can be implemented together in a single hardware device, and they can, for example, share memory, interfaces, and the like. The units can also be implemented in separate hardware components.

[0105] The term "one" here is not necessarily to be understood as restricting the number to exactly one element. Rather, it can also refer to multiple elements, such as two, three, or more. Similarly, every other counter used here should not be interpreted as restricting the number to the exact number stated. Instead, numerical deviations, both higher and lower, are possible unless otherwise specified.

[0106] The embodiments and features described for the object according to the first aspect apply accordingly to the objects proposed according to the second to fourth aspects and vice versa.

[0107] Other possible implementations of the invention also include combinations of features or embodiments described previously or subsequently with regard to the exemplary embodiments, even if not explicitly mentioned. In such cases, the person skilled in the art will also add individual aspects as improvements or additions to the respective basic form of the invention.

[0108] Further advantageous embodiments and aspects of the invention are the subject of the dependent claims and the exemplary embodiments of the invention described below. The invention will be explained in more detail below with reference to preferred embodiments and the accompanying figures. Fig. Figure 1 schematically shows a section of a photomask for microlithography with a defect in a structured coating according to one embodiment; Fig. Figure 2 shows a device for particle beam-induced processing of the defect in the photomask. Fig. 1 according to one embodiment; Fig. Figure 3 shows a flowchart of a method for determining a repair form for a defect in a photomask for microlithography according to one embodiment; Fig. 4 shows an image of the photomask from Fig. 1 together with a reference image and a difference image according to an embodiment; Fig. Figure 5 shows an enlarged view of a pre-repair form of the defect from the difference image. Fig. 4 according to one embodiment; Fig. Figure 6 illustrates the amount of curvature of a boundary line of the pre-repair mold. Fig. 5 together with a threshold value for the amount of curvature according to an embodiment; Fig. 7 shows a view similar to Fig. 5, wherein the pre-repair shape according to one embodiment is divided into a number of k pixels; Fig. Figure 8 shows a flowchart of a process for particle beam-induced processing of a defect in a photomask for microlithography; and Fig. Figure 9 shows a section of a test photomask for determining one or more scaling parameters that are used when correcting the pre-repair shape. Fig. 5 can be applied according to one embodiment.

[0109] In the figures, identical or functionally equivalent elements have been labelled with the same reference symbols, unless otherwise indicated. Furthermore, it should be noted that the representations in the figures are not necessarily to scale.

[0110] Fig. Figure 1 schematically shows a section of a photomask 100 for microlithography. In the example shown, the photomask 100 is a reflective photolithographic mask 100 for EUV lithography. The photomask 100 has a substrate 102. A surface of the substrate 102 and / or a coating of the substrate 102 is optically reflective, particularly at the wavelength with which the photomask 100 is exposed.

[0111] A structured coating 104 (pattern elements 104) is applied to the substrate 102. The coating 104 is, in particular, a coating made of an absorbing material. For example, one material of the coating 104 comprises a chromium layer. The thickness of the coating 104 is, for example, in the range of 50 nm to 100 nm. The structure size G of the structure formed by the coating 104 on the substrate 102 of the photomask 100 can vary at different positions of the photomask 100. For example, in the Fig. Figure 1 represents the width G of a region as the structure size. For example, the structure size G lies in a range of 20 to 200 nm. The structure size G can also be larger than 200 nm, for example, in the micrometer range.

[0112] In other examples, materials other than those mentioned can be used for the substrate and the coating. Furthermore, the photomask 100 can be a transmissive photomask instead of a reflective one. In this case, the substrate 102 is optically transparent at the wavelength used.

[0113] During the production of photomasks 100, occasional defects D may occur, for example, if etching processes do not proceed exactly as intended. Fig. Figure 1 shows such a defect D. In this example, it is a material excess because the coating 104 was not removed in one area of ​​the defect D. To ensure that a structure produced on a wafer in a lithography system with the photomask 100 has the desired shape and that the semiconductor device produced in this way fulfills the desired function, it is necessary to remove defects such as the one shown in Figure 1. Fig. To repair the defect shown in point D, or defects of other types. In the example of Fig. 1. It is necessary to remove the defect D in a targeted manner, for example by particle beam-induced etching.

[0114] Fig. Figure 2 shows a device 200 for particle beam-induced processing of a defect in a photomask for microlithography, such as the defect D of the photomask 100. Fig. 1. In Fig. Figure 2 schematically shows cross-sectional sections of some components of the device 200 that can be used for particle beam-induced repair, in this case etching, of the defect D of the photomask 100. Furthermore, the device 200 can also be used to image the photomask 100, in particular the structured coating 104 of the mask 100 and the defect D, before, during, and after the execution of a repair process.

[0115] The one in Fig. The apparatus 200 shown represents a modified scanning electron microscope 200. Here, a particle beam 202 in the form of an electron beam is used to repair the defect D. The use of an electron beam as the activating particle beam 202 has the advantage that the electron beam cannot damage the photomask 100, in particular its substrate 102, to a significant extent or at all.

[0116] In other examples, however, instead of an electron beam, another particle beam 202, e.g., an ion beam, atom beam, and / or molecular beam (not shown), can be used to activate a local chemical reaction. Furthermore, it is possible to use a laser beam in addition to the particle beam 202 to activate a local particle beam-induced repair process of the photomask 100 (not shown).

[0117] The device 200 is largely arranged in a vacuum housing 204, which is kept at a specific gas pressure by a vacuum pump 206.

[0118] For example, the device 200 is a repair tool for photomasks for microlithography, for example for photomasks for a DUV or EUV lithography system.

[0119] A photomask 100 to be processed is positioned on a sample stage 208. The sample stage 208 is, for example, designed to adjust the position of the photomask 100 in three spatial directions x, y, z and in three rotational axes around the three spatial directions x, y, z to an accuracy of a few nanometers.

[0120] The device 200 comprises a particle column 210 (e.g., an electron column). The particle column 210 includes a particle source 212 (e.g., an electron source) for providing the activating particle beam 202 (e.g., an electron beam). The particle column 210 also includes an electron or beam optic 214. The particle source 212 generates the particle beam 202, and the electron or beam optic 214 focuses the particle beam 202 and directs it at the exit of the column 210 onto the photomask 100. The particle column 210 also includes a deflection unit 216 (scan unit 216), which is configured to guide the particle beam 202 across the surface of the photomask 100, i.e., to scan or rasterize it.

[0121] The device 200 further comprises a detector 218 for detecting the secondary particles (e.g., secondary electrons and / or backscattered electrons) produced by the incident particle beam 202 at the photomask 200. The detector 218 is, for example, arranged in a ring around the particle beam 202 in the particle column 210, as shown. Alternatively and / or additionally to the detector 218, the device 200 can also have other / further detectors for detecting secondary particles (e.g., secondary electrons and / or backscattered electrons) (not shown).

[0122] The device 200 can also include one or more scanning probe microscopes, for example atomic force microscopes, which can be used to analyze the defect D of the photomask 100 (not shown).

[0123] The device 200 optionally includes a gas supply unit 220 for supplying process gas to the surface of the photomask 100. The gas supply unit 220 has, for example, a valve 222 and a gas line 224. The particle beam 202 directed from the particle column 210 to a location on the surface of the photomask 100 can, in conjunction with the process gas supplied from the outside via the valve 222 and the gas line 224 by the gas supply unit 220, perform a particle beam-induced processing process (e.g., electron beam-induced processing, EBIP). This includes, in particular, the deposition and / or etching of material.

[0124] The device 200 also includes a computing device 300, such as a computer. The computing device 300 is, in the example, of Fig. 2 arranged outside the vacuum housing 204. The computing device 300 can also be arranged remotely from the rest of the device 200, e.g. from the vacuum housing 204.

[0125] The computing device 300 includes, for example, a control unit 302 for determining a repair method 406 ( Fig. 5) of a defect D of a photomask 100 for microlithography. The control unit 302 has, for example, a provisioning unit 304 and several detection units 306 to 312.

[0126] The computing device 300 also includes, for example, a further control unit 314 for controlling the device 200. In particular, the further control unit 314 controls the provision of the particle beam 202 via the control of the particle column 210. For example, the further control unit 314 controls the scanning of the particle beam 202 across the surface of the photomask 100 via the control of the scan unit 216. The control unit 314 can optionally also control the provision of the process gas via the control of the gas supply unit 220.

[0127] The computing device 300 also includes, for example, an image generation unit 316. The image generation unit 316 receives, for example, measurement data from the detector 218 and / or other detectors of the device 200 and generates images from the measurement data that can be displayed on a monitor (not shown). Furthermore, images generated from the measurement data can be stored on a storage unit (not shown) of the computing device 300.

[0128] For the purpose of verifying the photomask 100 and in particular the structured coating 104 of the photomask 100, the device 200 is specifically designed to generate an image 400 of the photomask 100 from measurement data of the detector 218 and / or other detectors of the device 200 ( Fig. 1) or to capture an image 400 of a section of the photomask 100. The spatial resolution of the image 400 is, for example, in the range of a few nanometers.

[0129] The control unit 302 is configured to detect a defect D in the recorded image 400 ( Fig. 1) to detect and locate. Furthermore, the control unit 302 is configured to determine a geometric shape 402 of the defect D as a pre-repair shape 404 of the defect D. The determined geometric shape 402 of the defect D is, in particular, a two-dimensional geometric shape. The control unit 302 is also configured to correct the determined pre-repair shape 404 of the defect D in order to determine a repair shape 406 (final repair shape 406) of the defect D.

[0130] The following refers to Fig. 3 a method for determining a repair form 406 ( Fig. 5) of a defect D of a photomask 100 for microlithography described.

[0131] In a first step S1 of the procedure, an image 400 ( Fig. 1) at least part of the photomask 100 provided.

[0132] For example, the device 200 ( Fig. 2) a scanning electron microscope image 400 of a part of the photomask 100 is captured, in which a defect D of a structured coating 104 of the photomask 100 is depicted.

[0133] In a second step S2 of the procedure, a geometric shape 402 of the defect D is determined based on the image 400 as a pre-repair shape 404 with a boundary line 408.

[0134] For example, the geometric shape 402 is determined directly from the provided image 400. Alternatively, as in Fig. As shown in Figure 4, the geometric shape 402 of the defect D can also be determined in an image 412 derived from the provided image 400. For example, a reference image 410 is subtracted from the provided image 400, in which structures 104 of the photomask 100 and the defect D are depicted. The reference image 410 contains structures 104 similar to those in the provided image 400, but no defect D. The difference image 412 generated in this way contains only a representation of the defect D. In this case, the geometric shape 402 of the defect D in the difference image 412 can be determined as the pre-repair shape 404.

[0135] Determining the geometric shape 402, and thus the pre-repair shape 404, involves in particular the detection of edges of the depicted defect D in image 400 or image 412. Known edge detection algorithms can be applied for this purpose.

[0136] In the difference image 412 in Fig. Figure 4 shows an image of defect D with a circumferential border line 408.

[0137] In a third step S3 of the procedure, a curvature K1 of the boundary line 408 is determined at at least one position P1 of the boundary line 408, as shown in Fig. 5 illustrates.

[0138] Fig. Figure 5 shows, in particular, an enlarged view of the pre-repair shape 404 of defect D, determined in step S2. The boundary line 408 surrounding the pre-repair shape 404 has various curvatures, of which curvatures K1, K2, K3, and K4 are indicated with a reference numeral. It has been found that when using the pre-repair shape 404 for particle beam-induced treatment of defect D, the defect D is insufficiently repaired in boundary regions with a strong curvature K1 to K4. In particular, when using the uncorrected pre-repair shape 404 in convex regions Q1, Q2, and Q4 of the pre-repair shape 404 (i.e., in a positively curved section of the boundary line 408), an insufficient particle beam dose is applied. Furthermore, when using the uncorrected pre-repair shape 404 in concave areas Q3 of the pre-repair shape 404 (i.e., in a negatively curved section of the boundary line 408), an excessively high particle beam dose is applied.

[0139] To improve the processing of defect D of photomask 100, the present method ( Fig. 3) a curvature K1 to K4 of the edge line 408 of the pre-repair shape 404 is taken into account.

[0140] For example, in step S3, a mathematical magnitude C1 and a sign B1 of the curvature K1 are determined at at least one position P1 of the boundary line 408. For example, a magnitude C1 of the curvature K1 of the boundary line 408 at position P1 is determined based on a second derivative F'' ( Fig. 5) of the boundary line 408 is determined. In addition, for example, a sign B1 of the curvature K1 of the boundary line 408 at position P1 is also determined based on the second derivative F'' of the boundary line 408.

[0141] In Fig. Figure 5 shows by way of example that at four positions P1, P2, P3, P4 of the boundary line 408 a curvature K1, K2, K3, K4 is determined, whereby an magnitude C1, C2, C3, C4 and a sign B1, B2, B3, B4 of the curvature K1, K2, K3, K4 are determined in each case.

[0142] Determining the curvature K1 to K4 of the boundary line 408, for example, involves a comparison with a predetermined threshold value SW, as in Fig. 6 illustrates. Fig. Figure 6 shows a diagram of the magnitude C of a curvature as a function of a position P. For example, step S3 determines whether the magnitude C1 of the determined curvature K1 of the boundary line 408 at a given position P1 is greater than the predetermined threshold SW. Steps S4 and S5 are then performed for the given position P1, for example, only if step S3 determines that the magnitude C1 of the curvature K1 at this position P1 is greater than the predetermined threshold SW.

[0143] In a fourth step S4 of the procedure, at least one correction value X1 is determined for at least one position P1 of the boundary line 408 based on the determined curvature K1. For example, in step S4, the at least one correction value X1 is determined based on the determined magnitude C1 and the determined sign B1 of the curvature K1.

[0144] The at least one correction value X1 specifies, in particular, a displacement V1 of the boundary line 408 at the at least one position P1. In other words, the at least one correction value X1 specifies how the boundary line 408 of the pre-repair shape 404 should be displaced at the at least one position P1 in order to improve the particle beam-induced processing of the defect D in view of the curvature K1 at this position P1.

[0145] The displacement V1 of the boundary line 408 at the at least one position P1, which is defined by the correction value X1, is in particular a displacement of the boundary line 408 at position P1 in a direction substantially perpendicular to a tangent T1 to the boundary line 408 at this position P1.

[0146] In Fig. Figure 5 shows the respective determined correction values ​​X1, X2, X3, X4, the tangents T1, T2, T3, T4 and the displacements V1, V2, V3, V4 of the boundary line 408 corresponding to the correction values ​​for the four exemplary positions P1 to P4 of the boundary line 408.

[0147] For a position P1, P2, P4 of the boundary line 408 of the pre-repair shape 404, for which a positive curvature K1, K2, K4 (i.e., second derivative F'' > 0) is determined, the boundary line 408 is shifted at this position P1, P2, P4 such that the pre-repair shape 404 is enlarged at this point, as shown in Fig. Figure 5 illustrates this. Furthermore, for a position P3 of the boundary line 408 of the pre-repair shape 404, for which a negative curvature K3 (i.e. second derivative F'' < 0) is determined, the boundary line 408 is shifted at this position P3 so that the pre-repair shape 404 is reduced in size at this point P3.

[0148] Furthermore, the boundary line 408 is shifted at a position P1 to P4 where a curvature K1 to K4 of the boundary line has been determined (i.e., K1 to K4 > 0) and / or where the determined curvature K1 to K4 is greater than the predetermined threshold SW, by an amount that depends on the determined magnitude C1 to C4 of the curvature K1 to K4. In other words, the extent of the shift V1 to V4 is a function of the magnitude C1 to C4 of the curvature K1 to K4. Thus, the more curved the boundary line 408 is at a position P1 to P4, the greater the shift of the boundary line 408 at that position in order to correct the pre-repair shape 404 and thereby determine the repair shape 406.

[0149] In the example of Fig. For example, the boundary line 408 is more curved at position P2 than at position P1. Therefore, the magnitude B2 of the curvature K2 at position P2 is greater than the magnitude B1 of the curvature K1 at position P1. Consequently, the boundary line 408 is shifted more at position P2 (displacement V2, correction value X2) than at position P1 (displacement V1, correction value X1).

[0150] The at least one correction value X1 is determined in step S4, for example, based on the following equation: X1=A⋅B1⋅C1n.

[0151] In this equation, X1 denotes the correction value for position P1 of the boundary line 408, B1 denotes the sign of the curvature K1 at this position P1 (i.e., B1 can take the values ​​+1 or -1), and C1 denotes the magnitude of the curvature K1 at this position P1.

[0152] Furthermore, the equation above contains two optional scaling parameters A and n. The scaling parameter A is a scaling factor A (proportionality factor A). The scaling parameter n is a scaling exponent. The at least one correction value X1 can, for example, also be calculated solely based on the scaling factor A; in this case, only linear dependencies of the displacement V1 of boundary line 408 on the magnitude C1 of the curvature K1 are considered. If the correction value X1 is additionally calculated based on the scaling exponent n, non-linear dependencies of the displacement V1 of boundary line 408 on the magnitude C1 of the curvature can also be taken into account.

[0153] In a fifth step S5 of the procedure, a repair form 406 (final repair form 406) of defect D is determined. In particular, the pre-repair form 404 is corrected based on the at least one determined correction value X1 to determine the repair form 406 of defect D. Fig. Figure 5 shows the solid border line 408 of the pre-repair shape 404. Furthermore, displacements V1 to V4 of the pre-repair shape 404 are shown at positions P1 to P4, which lead to the corrected pre-repair shape, i.e., to the final repair shape 406 (in Fig. 5 shown with dashed lines).

[0154] As in Fig. As shown in Figure 7, the pre-repair shape 404 can be subdivided into a number of k pixels, 416. Fig. For the sake of clarity, only a few of the 416 pixels are labeled with a reference symbol. Fig. In addition, the boundary line 408 of the pre-repair shape 404 is shown in Figure 7. The pixels 416 of the pre-repair shape 404 that lie on the boundary line 408 (i.e., are touched and / or intersected by the boundary line 408) form boundary pixels 418 of the pre-repair shape 404. Some of the boundary pixels 418 are in Fig. 7 with a reference mark. As in Fig. As can be seen in Figure 7, the edge pixels 418 form a circumferential border of the pre-repair shape 404. If the pre-repair shape 404 is subdivided into pixels 416, then the at least one position P1 of the border line 408 of the pre-repair shape 404 corresponds to the position of at least one edge pixel 416. In this case, the at least one correction value X1 indicates how many pixels 420 are added to the pre-repair shape 404 or how many pixels 422 are removed from the pixels 416 of the pre-repair shape 404 to correct the pre-repair shape 404.

[0155] The following refers to Fig. 8 a method for particle beam-induced processing of a defect in a photomask 100 for microlithography is described.

[0156] In the first step S101 of the procedure according Fig. 8 will be a repair method 406 of defect D as in connection with the procedure according to Fig. 3 described and determined.

[0157] In a second step S102 of the procedure, the determined repair shape 406 is converted into a number of m pixels 416, 420 ( Fig. 5) subdivided.

[0158] In a third step S103 of the process, a particle beam 202 ( Fig. 2) A process gas is provided at each of the m pixels 416, 420 of the repair form 406 for processing the defect D. In step S103, a process gas can also be provided at each of the pixels 416, 420 of the repair form 406.

[0159] In the procedure according Fig. 8 can optionally include one or more scaling parameters A, n in a preceding step S100, which are used to determine at least one correction value X1 ( Fig. 5) are applied when determining the repair form 406, are determined empirically. For example, the empirical determination of one or more scaling parameters A, n involves analyzing a test photomask 500 ( Fig. 9) where the test photomask 500 has one or more defects D', D'' (similar to the defect D in Fig. 1) For example, image 502 (e.g., a SEM image) of at least part of the test photomask 500 is analyzed. For example, image 502 of the test photomask 500 is a difference image similar to image 412 in Fig. 4.

[0160] Analyzing image 502, or at least part of test photomask 500, serves in particular to develop a repair method (similar to repair method 406 in Fig. 5) to determine one or more defects D', D'' of the test photomask 500. A pre-repair form 504, 506 with a boundary line 508, 510 of the corresponding defect D', D'' is determined and corrected based on correction values, similar to in Fig. Figure 5 shows the pre-repair form 404 and the correction values ​​X1 to X4. The test photomask 500, for example, has one or more defects D', D'' with several concave and convex edge sections exhibiting different curvatures K, as shown in Fig. Figure 9 illustrates this (e.g., K has the values ​​-2; -1.5; -1; -0.5; +0.5; +1; +1.5; +2). The correction values ​​for the different curvatures K are given as in connection with Fig. As described in section 5, the correction values ​​for the various curvatures K are determined based on one or more scaling parameters A, n. A table of values ​​is tested for each of these scaling parameters. This means that repair methods for the one or more defects D', D'' of the test photomask 500 are used for several different values ​​of the one or more scaling parameters A, n.

[0161] Furthermore, the empirical determination of one or more scaling parameters A, n, for example, involves particle beam-induced processing of one or more defects D', D'' of the test photomask 500 (e.g., with the device 200). Fig. 2) In particular, the one or more defects D', D'' of the test photomask 500 are treated with the particle beam 202 based on the repair modes determined for the different values ​​of the one or more scaling parameters A, n, in order to assess the quality of a repair of the one or more defects D', D'' of the test photomask 500 as a function of the values ​​of the one or more scaling parameters A, n. Based on this, the optimal one or more scaling parameters A, n can then be determined.

[0162] In summary, the described procedure for determining repair form 406 ( Fig. 3), in which the curvature K1 of the edge line 408 of the pre-repair shape 404 is taken into account, an improved repair shape 406 for a defect D of a photomask 100 can be determined. This allows a dose of a particle beam 202 acting on a surface of the photomask 100 during the processing of the photomask 100 (method according to Fig. 8) can be adjusted more effectively. This allows for better processing and repair of a defect D in photomask 100. For example, an unwanted structure in the area of ​​defect D can be removed more effectively (e.g., completely). Or a missing structure in the area of ​​defect D can be added more effectively (e.g., completely). Furthermore, damage to adjacent structures (e.g., a capping layer) can be prevented and / or reduced (in the case of convex structures).

[0163] Although the present invention has been described using exemplary embodiments, it can be modified in many ways. REFERENCE MARK LIST 100 Photomasks 102 Substrat 104 Coating 200 Device 202 particle beam 204 cases 206 Pump 208 Sample table 210 particle column 212 Particle source 214 Electron / Beam Optics 216 Deflection unit 218 Detector 220 Gas Supply Unit 222 Valve 224 Gas pipeline 300 calculating device 302 Control unit 304 Deployment Unit 306-312 Investigation Unit 314 Control unit 316 Image generating unit 400 images Form 402 404 Pre-repair form 406 Repair form 408 Edge line 410 Reference image 412 Difference image 416 pixels 418 pixels 420 pixels 422 pixels 500 Test Photomasks 502 Image 504,506 Pre-repair form 508,510 Edge line A scaling parameter B1-B4 sign C, C1-C4 Amount D, D', D'' Defect F'' derivative G structural size K, K1-K4 curvature n scaling parameters P1-P4 Position Q1-Q4 area S1-S5 Process step S100-S103 Procedure step SW threshold T1-T3 tangent V1-V4 shift X1-X4 correction value x, y, z direction

Claims

[1] Method for determining a repair shape (406) of a defect (D) of a photomask (100) for microlithography, comprising the steps: a) Providing (S1) an image (400) or at least part of the photomask (100), b) Determine (S2) at least one section of a boundary line (408) of the defect (D) based on the image (400), c) Determining (S3) a curvature (K1) of at least one section of the boundary line (408) at at least one position (P1) of the at least one section of the boundary line (408), d) Determine (S4) at least one correction value (K1) for at least one position (P1) based on the determined curvature (K1), and e) Determine (S5) at least one corrected section of the boundary line by correcting at least one section of the boundary line (408) based on the at least one determined correction value (K1). [2] Method according to claim 1, wherein in step b) a geometric shape (402) of the defect (D) is determined based on the image (400) as a pre-repair shape (404) with a boundary line (408) that has at least one section of the boundary line (408), in step c) a curvature (K1) of the boundary line (408) is determined at at least one position (P1) of the boundary line (408), in step d) at least one correction value (K1) is determined for at least one position (P1) based on the determined curvature (K1), and In step e) the repair shape (406) is determined by correcting the pre-repair shape (404) based on at least one determined correction value (K1). [3] Method according to claim 1 or 2, wherein the at least one correction value (K1) specifies a displacement (V1) of the boundary line (408) at the at least one position (P1) in a direction substantially perpendicular to a tangent (T1) to the boundary line (408) at the at least one position (P1). [4] Method according to any one of claims 1 to 3, wherein in step c) an amount (C1) and a sign (B1) of the curvature (K1) of the boundary line (408) at at least one position (P1) of the boundary line (408) are determined, and in step d) the at least one correction value (K1) for the at least one position (P1) of the boundary line (408) is determined based on the determined amount (C1) and the determined sign (B1) of the curvature (K1). [5] Method according to any one of claims 2 to 4, wherein in step c) a second derivative (F'') of the boundary line (408) of the pre-repair shape (404) is determined, and an amount (C1) of the curvature (K1) of the boundary line (408) at the at least one position (P1) is determined based on the second derivative (F''). [6] Method according to any one of claims 2 to 5, wherein in step c) a second derivative (F'') of the boundary line (408) of the pre-repair shape (404) is determined, and a sign (B1) of the curvature (K1) of the boundary line (408) at the at least one position (P1) is determined based on the second derivative (F''). [7] Method according to any one of claims 2 to 6, wherein the repair shape (406) is determined in step e) such that for a position (P1) of the boundary line (408) of the pre-repair shape (404) for which a positive curvature (K1) is determined, the boundary line (408) at this position (P1) is shifted so that the pre-repair shape (404) at this position (P1) is enlarged, and For a position (P3) of the boundary line (408) of the pre-repair shape (404) for which a negative curvature (K3) is determined, the boundary line (408) at this position (P3) is shifted so that the pre-repair shape (404) at this position (P3) is reduced in size. [8] Method according to any one of claims 2 to 7, wherein in step c) an amount (C1) of the curvature (K1) of the boundary line (408) at at least one position (P1) is determined, and it is determined whether the amount (C1) of the curvature (K1) is greater than zero and / or greater than a predetermined threshold (SW), and the repair shape (406) in step e) is determined such that for a position (P1) of the boundary line (408) of the pre-repair shape (404), for which the determined amount (C1) of the curvature (K1) is greater than zero and / or greater than the predetermined threshold (SW), the boundary line (408) at this position (P1) is shifted by an amount which is a mathematical function of the amount (C1) of the curvature (K1) at this position (P1). [9] Method according to any one of claims 1 to 8, wherein the at least one correction value (K1) in step d) is determined based on one or more pre-determined scaling parameters (A, n) which specifies / specify a dependence of an extent of a displacement (V1) of the boundary line (408) at the at least one position (P1) on an amount (C1) of the determined curvature (K1) at the at least one position (P1). [10] Method according to claim 9, wherein one or more predetermined scaling parameters (A, n) include a predetermined scaling factor (A) and / or a predetermined scaling exponent (n). [11] Method according to any one of claims 2 to 10, wherein the determined pre-repair shape (404) is subdivided into a number of k pixels (416), the pixels (416) of the pre-repair shape (404) that are touched and / or intersected by the border line (408) form border pixels (418) of the pre-repair shape (404), which corresponds to at least one position (P1) of the border line (408) of a position (P1) of at least one border pixel (418), and the repair shape (406) in step e) is corrected by adding or removing, adjacent to at least one edge pixel (418), one or more pixels (420, 422) to or from the pre-repair shape (404) based on at least one determined correction value (K1). [12] Method according to claim 11, wherein in step e) for a boundary pixel (418) for which a positive curvature (K1) is determined, one or more further pixels (420) are added in a direction substantially perpendicular to a tangent (T1) to the boundary line (408) at this boundary pixel (418) to the pre-repair shape (404), and for a boundary pixel (418, 422) for which a negative curvature (K3) is determined, this boundary pixel (418, 422) and / or other neighboring pixels (422) are removed in a direction substantially perpendicular to a tangent (T3) to the boundary line (408) at this boundary pixel (418) from the pre-repair shape (404). [13] Method for particle beam-induced processing of a defect (D) of a photomask (100) for microlithography, comprising Determining (S101) a repair method (406) of the defect (D) according to one of claims 1 to 12, Subdividing (S102) the determined repair shape (406) into a number of m pixels, and Providing (S103) a particle beam (202) at each of the m pixels of the repair shape (406) to process the defect (D). [14] Method according to claim 13 and claim 9 or 10, wherein one or more scaling parameters (A, n) are / are empirically determined by analyzing and processing a test photomask (500). [15] Computer program product comprising instructions which, when the program is executed by at least one computer, cause it to execute the method according to any one of claims 1 to 14. [16] Control device (302) for determining a repair method (406) for a defect (D) of a photomask (100) for microlithography, comprising a provisioning unit (304) for providing an image (400) or at least a part of the photomask (100), a first investigative unit (306) to determine at least one section of a boundary line (408) of the defect (D) based on the image (400), a second investigation unit (308) for determining a curvature (K1) of at least one section of the boundary line (408) at at least one position (P1) of the at least one section of the boundary line (408), a third determination unit (310) for determining at least one correction value (K1) for at least one position (P1) based on the determined curvature (K1), and a fourth investigation unit (312) for determining at least one corrected section of the boundary line by correcting the at least one section of the boundary line based on the at least one determined correction value (K1).

Images