Method for inspecting microlithographic photomasks, computer program product, system for inspecting a microlithographic photomask, method for repairing a microlithographic photomask and method of microlithography

Abstract

Method for inspecting a microlithographic photomask (17), wherein with the aid of a mask inspection device comprising an illumination lens (16) and a projection lens (22), the image of a photomask (17) illuminated by means of the illumination lens (16) is projected by the projection lens (22) onto an image sensor (24) of a camera (23) arranged in the image plane of the projection lens (22). The photomask (17) is placed on a positioning system (26) that is designed to displace the photomask (17). The photomask (17) is displaced using the positioning system (26) such that the image sensor (24) captures a first image line (101) that corresponds to a first region on the photomask (17). The photomask (17) is displaced using the positioning system (26) such that the image sensor (24) captures a second image line (102) that corresponds to a second region on the photomask (17). There is an overlap (200) between the first region and the second region such that the first and the second image line (101, 102) comprise a piece of image information regarding a structure of the photomask (17) arranged within the overlap (200). A defect detection method is applied to the piece of image information in the first image line (101) in order to determine a first piece of defect information. A defect detection method is applied to the piece of image information in the second image line (102) in order to determine a second piece of defect information. The first piece of defect information is compared with the second piece of defect information in order to determine a piece of information regarding the correctness of the structure of the photomask (17) arranged within the overlap (200). The invention also relates to a system for inspecting a microlithographic photomask, a computer program product, a method for repairing a microlithographic photomask and a method of microlithography.

Description

CZVS 151PW028 . 10 . 2025 / PHMethod for inspecting mi cr ©lithographic photomasks , computer program product , system for inspecting a mi cr ©lithographic photomask , method for repairing a mi cr ©lithographic photomask and method of microlithography

[0001] The invention relates to a method for inspecting a mi- crolithographic photomask . The invention also relates to a computer program product , a system for inspecting a microlith- ographic photomask, a method for repairing a microlithographic photomask and a method of microlithography .

[0002] Photomasks are used in microlithographic proj ection exposure apparatuses used to produce components such as integrated circuits with particularly small structures . The photomask illuminated by very short-wave extreme ultraviolet radiation (EUV radiation) is imaged onto a lithography obj ect in order to trans fer the mask structure to the lithography ob- j ect .

[0003] To ensure a high quality of the image generated on the lithography obj ect , it is necessary for the photomask to be true to si ze and not adversely af fected by structural defects . It is known practice to subj ect photomasks to an inspection, either prior to operation in a microlithographic proj ection exposure apparatus or during an interruption of operation . To this end, what is known as an aerial image of a portion of the photomask is created using a mask inspection system, the photomask in the process being imaged not on a lithography obj ect but on an image sensor of an EUV camera . Using the imaging onto the image sensor as a basis , it is possible to use a defect detection method to make an assessment as to whether the photomask is without defects . Defect detection methods are known in the prior art .

[0004] For example , a defect may consist in a structure part of the photomask or the distance between two structure parts not being suf ficiently wide . Additionally, a structure part may comprise a locali zed defect not provided for in the envisaged design of the photomask or an unwanted "bridge" to an adj acent structure part . During an exposure of a mask comprising a mask defect , the defect would be imaged on the light-sensitive layer as an irregular shadow or bright region and, following the subsequent processing steps , would also be found in the component ultimately produced . Various repair methods that can be used to remove defects or at least reduce these to a harmless level are known from the prior art . For this purpose , a material of the photomask is typically removed, added and / or structurally modi fied during a repair step . To this end, an electron beam or an ion beam may be used to influence the photomask, for example . Relatively long straight defects may also be removed by mechanical action ( scratching) . The use of a laser is also possible but usually not precise enough .

[0005] Any undetected defect on a photomask left unrepaired may render an entire batch of components produced unusable . Hence , it is desirable that preferably no defect remains undetected during the inspection of microlithographic photomasks so that all defects actually present can be repaired .

[0006] Defect detection methods may ascertain a piece of defect information that comprises one or more defect candidates . A defect candidate is a detected potential defect in the structure of the photomask . Defect detection methods applied during a mask inspection typically have high sensitivity and low speci ficity . The sensitivity is the probability that an actual defect is in fact present at a detected defect candidate . The speci ficity is the probability that no actual defect is in fact present should no defect candidate be detected . Accordingly, the defect detection methods are typically designedsuch that all actual defects present are detected while accepting a few detected defect candidates at which no actual defect is present . In other words , the applied defect detection methods typically have a low " false negative rate" and a high " false positive rate" . As explained above , the low false negative rate is desired so that preferably no actual defects present remain undetected . However, the high false positive rate is disadvantageous because it is typically a human who performs the last assessment of the detected defect candidates before the photomask is repaired, and so the human could be presented with numerous defect candidates where no actual defect on the photomask is present ; this is time-consuming and expensive .

[0007] The problem addressed by the invention is that of presenting a method and a computer program product for inspecting microlithographic photomasks , in which the aforementioned disadvantages are reduced . The problem is solved by the features of the independent claims . Advantageous embodiments are specified in the dependent claims .

[0008] In a method according to the invention for inspecting a microlithographic photomask and with the aid of a mask inspection device comprising an illumination lens and a proj ection lens , the image of a photomask illuminated by means of the illumination lens is proj ected by the proj ection lens onto an image sensor of a camera arranged in the image plane of the proj ection lens . The photomask is placed on a positioning system that is designed to displace the photomask . The photomask is displaced using the positioning system such that the image sensor captures a first image line that corresponds to a first region on the photomask . The photomask is then displaced using the positioning system such that the image sensor captures a second image line that corresponds to a second region on thephotomask, with there being an overlap between the first region and the second region such that the first and the second image line comprise a piece of image information regarding a structure of the photomask arranged within the overlap . A defect detection method is applied to the piece of image information in the first image line in order to determine a first piece of defect information from the overlap . A defect detection method is applied to the piece of image information in the second image line in order to determine a second piece of defect information from the overlap . The first piece of defect information and the second piece of defect information are compared with each other in order to determine a piece of information regarding the correctness of the structure of the photomask arranged within the overlap .

[0009] The invention has recogni zed that a piece of image information regarding the structure of the photomask arranged within the overlap is captured twice in the case of two overlapping image lines , and so , by way of the captured image lines , a piece of information regarding a defect of the structure of the photomask arranged within the overlap is also present twice . The invention thus proposes to apply a defect detection method to the piece of image information regarding the structure of the photomask arranged within the overlap found in both image lines in order to determine two independent pieces of defect information by way of the region of the overlap on the photomask, and these two independent pieces of defect information are subsequently compared with each other .

[0010] In the case of a mask inspection device , the image field on the surface of the photomask that corresponds to the area of the image sensor is small in comparison with the area of the photomask . During an exposure procedure , the photomask is moved relative to the image sensor by way of a scanning movement of the positioning system . This makes it possible torecord an image line extending over the length of the photomask during a continuous exposure procedure . In order to create an overall image of the photomask, multiple successively recorded image lines typically are aligned relative to each other on the basis of the structure of the photomask arranged within the overlap and are stitched together .

[0011] The positioning system may be designed to displace the photomask in an X- and / or a Y-direction . The first image line may be captured by virtue of the positioning system displacing the photomask in the X-direction . The second image line may also be captured by virtue of the positioning system displacing the photomask in the X-direction . Before the second image line is captured, the positioning system can displace the photomask in the Y-direction . Multiple image lines may also be captured by virtue of the photomask being displaced in opposite directions by way of the positioning system . Thus , the X- direction might be positive for the first image line and negative for the second image line . The opposite might also be the case . The extent of the overlap of the image lines or the overlap of the regions on the photomask corresponding to the captured image lines may be set by way of the displacement in the Y-direction .

[0012] In principle , the extent of the overlap between the first and the second region on the photomask may be as desired . All that is essential to the invention is that there is an overlap between the first and the second region . In other words , the overlap parameter is greater than 0 ( 0% ) and less than or equal to 1 ( 100% ) . The overlap parameter relates to the relationship between the overlap and an image line . In other words , the overlap parameter states the extent to which an image line is encompassed by the overlap . The overlap parameter may be less than 0 . 1 ( 10% ) , preferably less than 0 . 05 ( 5% ) and further preferably less than 0 . 02 ( 2 % ) . Such smalloverlap parameters are advantageous in that the entire photomask can be captured quicker. For an overlap parameter of less than 1, the photomask needs to be displaced in the Y-direction before the second image line is captured. The greater the overlap, the more image lines need to be captured overall in order to capture the entire photomask, and so the overall capture of the photomask requires more time. However, the greater the overlap, the more regions of the photomask are also captured at least twice in an image line, and so a two-fold overlap is present for more regions. For a complete two-fold overlap of the photomask, the overlap parameter may be between 1 / 2 (50%) and 2 / 3 (approx. 66.6%) for example. An overlap parameter of 1 / 2 is advantageous in that each region of the photomask is just captured exactly two times in an image line. It is preferable for the overlap parameter to be adjustable by an operator such that speed can be balanced against accuracy.

[0013] The steps of displacing the photomask may be repeated multiple times in order to capture a plurality of pairs of first and second image lines. It is then possible to specify a rate, on the basis of which the steps of applying the defect detection method and comparing the pieces of defect information are performed for only every x-th pair of image lines. In other words, the specified rate is used to determine the piece of information regarding the correctness for only a subset of the plurality of pairs of first and second image lines. As a result, the frequency of determining the piece of information regarding the correctness may be specified. For example, the information regarding the correctness might be determined only for every 2nd or 3rd pair of captured image lines, reducing the determination outlay to a half or a third, and thus offering a trade off between speed and reliability

[0014] A specified region may be provided; an operator may use this to specify a certain region on the photomask that shouldbe captured at least twice by means of an overlap . This is advantageous in that it is only the determined region and not the entire photomask that is captured twice , of fering a balance between time outlay and accuracy . In an embodiment , the overlap region is fixedly predetermined on the basis of structures to be registered .

[0015] The piece of information determined by the method according to the invention and regarding the correctness may be compared with an expected correctness of the structure of the photomask arranged within the overlap . A piece of calibration information for the mask inspection device may be determined on the basis of such a comparison . It is also possible that a piece of calibration information for a defect detection method is determined on the basis of the comparison . The expected correctness ( the so-called " ground truth" ) may emerge from a reference photomask being available , the latter being prepared with structural defects in advance ( a so-called "defect programmed mask" , DPM) . In other words , the method according to the invention may be performed using the reference photomask, and the result may be compared with the ground truth . The reference photomask may also be a photomask with production quality confirmed in advance ( a so-called "production-grade mask" ) , for which the assumption may be made that it is actually defect- free and thus has no structural defects . The piece of calibration information may be used to set or readj ust the mask inspection device and / or a defect detection method, in order to remove or at least reduce potential deviations from the ground truth . In other words , the piece of calibration information can preferably be used to reduce a deviation between the determined piece of information regarding the correctness and the correctness to be expected . An advantage thereof lies in a simpli fied calibration on the basis of a few captured image lines .

[0016] A piece of defect information may comprise a defect candidate should a defect detection method have detected a potential defect . Defect candidates may be presented to a human observer for a final assessment , in order to veri fy the defect candidates as actual defects or discard these as a false positive . Veri fied defect candidates that should represent actual defects in the structure of the photomask may be repaired in a subsequent repair step .

[0017] The piece of image information in the captured image lines may be disturbed in various ways on account of di f ferent random processes . For example , the piece of image information may be disturbed by superimposed noise . In other words , the image information may be superimposed by a first reali zation of a random process in the first image line and by a second reali zation of the random process in the second image line . On account of the superimposition of random processes , a defect detection method may therefore result in di f ferent results for the piece of image information in the first image line and the piece of image information in the second image line . The ratio between values of the piece of information and a random process is frequently referred to as " signal-to-noise ratio" ( SNR) . For example , a low SNR may speci fy that the piece of image information is very noisy .

[0018] It is also possible that the piece of image information in the first image line and the piece of image information in the second image line di f fer from each other on account of optical ef fects , which may for example be based on optical scattering . The various optical ef fects may also lead to a defect detection method resulting in di f ferent results for the piece of image information in the first image line and the piece of image information in the second image line .

[0019] A di f ferent result means that a speci fic defect detection method for example ascertains a piece of defect information with a defect candidate in the piece of image information of the first image line while the same defect detection method does not ascertain any defect information with a defect candidate in the piece of image information of the second image line .

[0020] The first defect detection method and the second defect detection method may be the same defect detection method . However, it is also possible that the first and second defect detection methods di f fer . The advantages and disadvantages of various speci fic defect detection methods may then be combined with or balanced against each other .

[0021] By virtue of the pieces of defect information determined by a defect detection method being compared with one another according to the invention, the reliability of the applied defect detection methods is increased . A potential defect of the structure of the photomask arranged within the overlap may thereby be correctly detected with increased reliability . For example , i f the first piece of defect information contains a defect candidate , i . e . the defect detection method has detected a defect in the piece of image information regarding the structure of the photomask in the first image line arranged within the overlap, then the defect candidate may be veri fied by means of the second piece of defect information . For example , it is possible that a defect candidate is only veri fied as a defect actually present i f it is contained both in the first and in the second piece of defect information . By virtue of the first piece of defect information and the second piece of defect information being compared with one another, it is possible - in other words - to veri fy a defect candidate in the first piece of defect information using a defect candidate in the second piece of defect information . In that case ,the piece of information regarding the correctness of the structure of the photomask arranged within the overlap can thus state that a defect is present i f a speci fic defect candidate is present both in the first piece of defect information and in the second piece of defect information ( " logical AND" ) .

[0022] Whether a defect candidate in the first piece of defect information corresponds to a defect candidate in the second piece of defect information may be based on a piece of identity information . The piece of identity information may state that two defect candidates must have the same centre or must be arranged within a certain radius on the photomask in order to be considered identical . The piece of identity information may likewise state that the shape of two defect candidates must be identical or must be within a certain shape tolerance . It is preferable for an operator to be able to speci fy the piece of identity information .

[0023] By virtue of the determined pieces of defect information being compared with each other according to the invention, the susceptibility of applied defect detection methods to a superimposed random process or optical ef fects is also reduced . For example , should a random process such as signi ficant noise be superimposed on the piece of image information in the captured image line , the case may arise where a defect detection method does not detect an actually present defect despite the high sensitivity . In such a scenario , it is preferable for a defect to be already veri fied as such i f merely one piece of defect information contains a speci fic defect candidate . In other words , in that case the piece of information regarding the correctness of the structure of the photomask arranged within the overlap may thus state that a defect is present i f a certain defect candidate is present in atleast one of the first and the second piece of defect information ( " logical OR" ) .

[0024] I f none of the pieces of defect information comprise a defect candidate , the piece of information regarding the correctness of the structure of the photomask arranged within the overlap may state that the structure of the photomask arranged within the overlap is defect free .

[0025] The piece of information regarding the correctness of the structure of the photomask arranged within the overlap may preferably be output . For example , the piece of information regarding the correctness may be communicated or transmitted to a human user or a computer . To this end, the piece of information regarding the correctness of the structure of the photomask arranged within the overlap may be for example displayed on a display - in particular as text or an image - or output by way of a loudspeaker . It is also possible that the piece of information regarding the correctness is communicated - in particular to a computer - in a manner that is not readable by a human . An output in the form of a database entry or as a text file ( a so-called " log file" ) is also possible .

[0026] The piece of information regarding the correctness of the structure of the photomask arranged within the overlap may comprise a piece of reliability information . A piece of reliability information speci fies the extent to which the result is classi fied as reliable or certain . For example , a result may be classi fied as very reliable should it have been veri fied on the basis of the first and the second piece of defect information . In other words , a detected defect may be classi fied as very reliable should a speci fic defect candidate be contained in the first and in the second piece of defect information . A detected defect may be classi fied as less reliable should a certain defect candidate be contained in only one of the firstand second pieces of defect information . The piece of reliability information may comprise a number that comprises the number of defect candidates that veri fy the corresponding defect . However, it is also possible that the piece of reliability information comprises the number of non-corresponding pieces of defect information . The piece of reliability information may also be speci fied in per cent or as a decimal number . The piece of reliability information may preferably be output instead of , or together with, the piece of information regarding the correctness . The advantage thereof is that an assessment of the reliability of the mask inspection is rendered available . In that case , a final human assessment may for example concentrate on the detected defects for which a low reliability was output . In other words , the human evaluation may then be concentrated on detected defects for which the method according to the invention has ascertained a less reliable result .

[0027] Should the piece of reliability information speci fy that a result is classi fied as having low reliability, further steps may be undertaken in order to improve the reliability of the result . For example , the method may be repeated using other defect detection methods that might lead to a more reliable result . Additionally, settings of the mask inspection device may be modi fied in response to the piece of reliability information . For example , the image lines might be captured more slowly, as this typically leads to a higher and hence better SNR . The illumination of the photomask might also be increased, as this typically also leads to a higher SNR .

[0028] A defect detection method may comprise the application of a trained algorithm . The algorithm may comprise a model of machine learning . Training, validating and applying models of machine learning are known from the prior art . In particular, models comprising arti ficial neural networks , which form thebasis of what is known as "deep learning" , are known . The algorithm may be trained on the basis of training data . The training data may contain a plurality of pairs , each comprising an image line and a piece of defect information ( so-called " supervised learning" ) . It is also possible that only sections of image lines are used in the training data . The image lines in the training data may additionally also comprise various random processes such that a piece of image information in the image lines is superimposed by one or more random processes , whereby the trained algorithm may be less susceptible to certain random processes . The input of the trained algorithm may comprise an image line or a section of an image line . The output of the trained algorithm may comprise a piece of defect information . It is also possible that the step of comparing the first piece of defect information with the second piece of defect information is performed by the trained algorithm . In that case , the output of the trained algorithm may comprise a piece of information regarding the correctness of the structure of the photomask arranged within the overlap, while the input comprises a plurality of image lines .

[0029] On the basis of the associated pieces of reliability information, results may be added to the training data . For example , results classi fied as very reliable may be added to the training data . This is advantageous in that the algorithm may be trained further on the basis of results classi fied as reliable . Future applications of the corresponding defect detection method may be further improved thereby without a human needing to select further training data manually .

[0030] The photomask may also be displaced in such a way using the positioning system that the image sensor captures a third image line that corresponds to a third region on the photomask . In that case , there may be an overlap between the first region and the third region and also between the second regionand the third region. In other words, a simultaneous overlap of the first, second and third regions is present in that case, i.e. a threefold overlap. In that case, the first, the second and the third image line comprise a piece of image information regarding a structure of the photomask arranged within the overlap. In other words, the piece of image information is available three times in that case. A third defect detection method may also be applied to the third image line in that case, in order to determine a third piece of defect information. The first piece of defect information, the second piece of defect information and the third piece of defect information may be compared with one another in order to determine the piece of information regarding the correctness of the structure of the photomask arranged within the overlap. The explanations given above in relation to the examples with two image lines or two pieces of defect information may naturally be applied to three image lines or three pieces of defect information. For example, the piece of reliability information may in that case obtain the highest value should all three pieces of defect information contain a defect candidate, i.e. should the logical AND thus be confirmed by all three pieces of image information. The pairs of training data may also comprise three image lines, etc.

[0031] Should the comparison according to the invention be based on three pieces of defect information, it is moreover possible to refine the decision-making process even further during the verification of a defect candidate. For example, a defect candidate may be verified as an actual defect if it is present in all three pieces of defect information, i.e. all three pieces of defect information fulfil the logical AND. It is also possible that a defect candidate is already verified as an actual defect if it is encompassed by a plurality of the pieces of defect information - i.e. by at least two pieces of defect information in the case of a threefold overlap.

[0032] A complete threefold overlap of the photomask may be obtained if the overlap parameter is between 2 / 3 and 3 / 4 (75%) . In other words, each region of the photomask may be captured at least three times in an image line in that case. A partial threefold overlap may already arise above an overlap of more than 1 / 2.

[0033] A person skilled in the art knows that fourfold overlaps and overlaps of even higher order are possible in the event of a corresponding overlap parameter. Accordingly, the exemplary embodiments based on a twofold or threefold overlap may also be applied accordingly to a fourfold overlap or an overlap of even higher order. In one embodiment, the photomask is repeatedly captured in full such that an even greater number of image lines, which are available for the comparison according to the invention of pieces of defect information, is captured .

[0034] If the overlap parameter is 1 (100%) , then a certain region of the photomask is repeatedly captured in image lines in full. In that case, there is no displacement of the photomask in the Y-direction by way of the positioning system. In that case, the captured image lines fundamentally only differ in terms of superimposed random processes and possible optical effects, for example on account of minimal displacements between components of the mask inspection device or the photomask.

[0035] In an embodiment, multiple image lines are superimposed on one another and combined (e.g. added or averaged) in order to create a combined image line. If superimposed random processes are uncorrelated over multiple image lines, i.e. individual realizations of random process are not correlated, then this has the advantage of an improved SNR of the combined im-age line . The method according to the invention may then preferably be performed on the basis of combined image lines . This in turn is advantageous in that it is possible to use defect detection methods that are less susceptible to superimposed random processes . Additionally, the piece of identity information may be set in a manner adapted to combined image lines . In particular, a combined image line may be based on image lines that were captured with an overlap parameter of 1 .

[0036] In principle , the invention may also be applied in the case of photomasks that are designed to be illuminated by "deep ultraviolet light" ( DUV radiation) .

[0037] The invention also relates to a computer program product or a set of computer program products comprising program parts which, when loaded into a computer or into networked computers connected to a device according to the invention, are designed to perform the method according to the invention .

[0038] Furthermore , the invention also relates to a method for repairing a microlithographic photomask, wherein a material of the photomask is removed, added and / or structurally modi fied during a repair step . The repair step is preceded by determining a piece of information regarding the correctness of the photomask by means of a mask inspection method according to the invention . The repair step is subsequently performed in a region of the photomask with an insuf ficient correctness .

[0039] Furthermore , the invention also relates to a method of microlithography, wherein with the aid of a proj ection exposure apparatus comprising an illumination lens and a proj ection lens , the image of a photomask illuminated by means of the illumination lens is proj ected by the proj ection lens onto a substrate coated with a light-sensitive layer and arranged in the image plane of the proj ection lens . The exposure of thesubstrate is preceded by determining a piece of information regarding the correctness of the photomask by means of a mask inspection method according to the invention. The exposure of the substrate is performed only if sufficient correctness is determined .

[0040] A region in which one or more defects were verified may be insufficiently defect free.

[0041] Furthermore, the invention also relates to a system comprising a mask inspection device and a computer. The computer is configured to perform a mask inspection method according to the invention.

[0042] The disclosure encompasses development of the method with features that are described in the context of the system according to the invention. The disclosure encompasses developments of the system which are described in the context of the method according to the invention.

[0043] The invention is described by way of example below on the basis of advantageous embodiments with reference to the accompanying drawings, in which:Fig. 1: shows a schematic illustration of a mask inspection device;Fig. 2: shows a schematic illustration of a capture according to the invention of two image lines on a photomask;Fig. 3: shows a schematic illustration of a twofold overlap according to the invention; andFig. 4: shows a schematic illustration of a threefold overlap according to the invention.

[0044] Microlithographic photomasks 17 may be examined by means of a mask inspection device shown in Fig. 1.

[0045] In general, microlithographic photomasks 17 are intended to be used in a microlithographic projection exposure apparatus (not illustrated) . In the microlithographic projection exposure device, the photomask 17 is illuminated with extreme ultraviolet radiation (EUV radiation) at a wavelength of for example 13.5 nm in order to image a structure formed on the photomask 17 onto the surface of a lithographic object in the form of a wafer. The wafer is coated with a photoresist that reacts to the EUV radiation. The mask inspection device is used to examine whether the photomask meets the specifications and is free from contaminations.

[0046] In what is known as DUV lithography, corresponding projection exposure apparatuses and mask inspection devices are illuminated with DUV radiation at a wavelength of 193 nm to 248 nm, for example.

[0047] According to Fig. 1, the photomask 17 is arranged in the mask inspection device such that an EUV beam path 15 emanating from an EUV radiation source 14 is guided via an illumination lens 16 onto the photomask 17. The illumination lens 16 is used to shape the EUV radiation to form a beam used to illuminate, with uniform brightness, an examination field on the surface of the photomask 17. The illuminated region may have dimensions of 0.5 mm x 0.8 mm, for example. A field stop used to delimit the illuminated region to the examination field on the surface of the photomask 17 is arranged in the illumination lens 16. Using a positioning system 26, it is possible to move the photomask in the XY-plane in order to bring different examination fields on the surface of the photomask 17 into the region of the EUV beam path 15.

[0048] The edge lengths of the photomask 17 may be between 100 mm and 200 mm, for example. The photomask may have an aspect ratio of between 1:1 and 1:3, preferably between 1:1 and 1:2 and particularly preferably of 1:1 or 1:2. The photomask may be configured to be substantially rectangular. The photomask may preferably have a length and a width of 5 to 7 inches (12.7 cm to 17.8 cm) , particularly preferably a length and a width of 6 inches (15.2 cm) . In an alternative thereto, the photomask may have a length of 5 to 7 inches (12.7 cm to 17.8 cm) and a width of 10 to 14 inches (25.4 cm to 35.6 cm) , preferably a length of 6 inches (15.2 cm) and a width of 12 inches (30.5 cm) .

[0049] The EUV beam path 15 reflected off the photomask 17 continues through a projection lens 22 to an EUV camera 23, which is equipped with an image sensor 24. The projection lens 22 is used to image the examination field on the surface of the photomask 17 onto the image sensor 24 of the EUV camera 23. The EUV radiation source 14, the illumination lens 16, the photomask 17, the projection lens 22 and the EUV camera 23 are arranged in a vacuum housing 40 that is under negative pressure during the operation of the mask inspection device.

[0050] The EUV radiation source 14 is a plasma radiation source, in which the EUV radiation is emitted from a plasma at a wavelength of 13.5 nm. Tin is a medium that can be used to generate a plasma suitable for emitting such EUV radiation. A laser beam can be made to impinge on a droplet of the medium for the purpose of creating the plasma.

[0051] The illumination lens 16 and the projection lens 22 may comprise mirrors off which the EUV radiation is reflected. The mirrors may be designed as EUV mirrors which have a particularly high reflectivity for EUV radiation. The optical surfaceof the EUV mirrors may be formed by a highly reflective coating . This may be a multilayer coating, in particular a multilayer coating having alternating layers of molybdenum and silicon . Using such a coating, it is possible to reflect approximately 70% of the incident EUV radiation .

[0052] The proj ection lens 22 has a magni fication factor of more than 100 . In order to be able to record the entirety of the image generated by the examination field on the surface of the photomask 17 , the area of the image sensor 24 is greater than the area of the examination field in accordance with the magni fication factor . The image sensor 24 may have dimensions of the order of 100 mm to 200 mm, for example . The image sensor 24 comprises a multiplicity of parallel pixel lines , which span a pixel array . The image sensor 24 is aligned such that the longitudinal direction of the pixel lines 36 corresponds to the X-direction . In the event of a movement of the photomask 17 in the X-direction, the image of the photomask 17 on the image sensor 24 moves parallel to the longitudinal direction of the pixel lines , whereby an image line may be captured .

[0053] A first image line 101 is accordingly captured by virtue of the positioning system 26 displacing the photomask 17 in the positive X-direction . The first image line 101 comprises a piece of image information regarding the region on the photomask 17 illuminated during the displacement . A second image line 102 is also captured by virtue of the positioning system 26 displacing the photomask in the X-direction, albeit in the negative X-direction in this case . The second image line 102 also comprises a piece of image information regarding the region on the photomask 17 illuminated during the displacement . Before the second image line 102 is captured, the positioning system 26 displaces the photomask 17 in the positive Y-direction . As illustrated schematically in Fig . 2 , thephotomask 17 is displaced in the Y-direction in such a way that there is an overlap between the regions on the photomask 17 that in each case correspond to the first image line 101 and the second image line 102 ; this is indicated by the dashed lines .

[0054] Fig . 3 shows the first image line 101 and the second image line 102 in corresponding fashion on the photomask 17 . There is a twofold overlap 200 between the two image lines 101 , 102 , since said overlap results from the overlap of two image lines . On account of the twofold overlap 200 , the first and second image lines 101 , 102 comprise a redundant piece of image information regarding the region on the photomask 17 that corresponds to the overlap . In other words , the same piece of image information is contained once in the first image line 101 and once in the second image line 102 .

[0055] The overlap parameter is approximately 0 . 3 in the example of Fig . 3 . The overlap parameter was set by an operator of the mask inspection device before the capture of the image lines 101 , 102 . More precisely, the operator has speci fied the region of twofold overlap 200 as a speci fied region in this example such that only this speci fic region and not the photomask overall is captured twice .

[0056] A defect detection method is applied to the piece of image information in the first image line 101 in order to determine a first piece of defect information . The first piece of defect information allows conclusions to be drawn as to whether the region on the photomask 17 that corresponds to the overlap is free from defects . The same defect detection method is applied to the piece of image information in the second image line in order to determine a second piece of defect information . The second piece of defect information also allowsconclusions to be drawn as to whether the region on the photomask 17 that corresponds to the overlap is free from defects .

[0057] The first piece of defect information and the second piece of defect information are compared with each other in order to determine a piece of information regarding the correctness of the structure of the photomask arranged within the overlap . The two pieces of defect information are thus combined with each other in order to obtain in-depth information regarding the correctness . In an example , the first piece of defect information comprises a defect candidate . The latter represents a defect on the photomask 17 detected by the defect detection method . I f the second piece of defect information also contains the defect candidate , i . e . i f the defect detection method detects the same defect on the photomask 17 in both image lines , then the defect candidate is veri fied as an actual defect . The actual defect is subsequently repaired in a repair step . I f the second piece of defect information does not contain the defect candidate , then the defect candidate is discarded as a false positive in this example .

[0058] The piece of image information in the captured image lines is disturbed by a superimposed noise . The piece of image information in the first image line 101 thus di f fers from the piece of image information in the second image line 102 in that it is disturbed by a di f ferent reali zation of the superimposed noise . Since the first image line 101 is captured by virtue of the photomask 17 being displaced in the positive X- direction and the second image line 102 is captured by virtue of the photomask 17 being displaced in the negative X-direc- tion, the piece of image information in the first image line 101 also di f fers from the piece of image information in the second image line 102 in terms of optical ef fects that are based on minimal displacement between the individual components of the mask inspection device .

[0059] In an example , the superimposed noise leads to the defect detection method identi fying a defect candidate in the first image line 101 , while the defect detection method does not detect the defect candidate in the second image line 102 . However, since a defect detection method was designed with a low " false negative rate" , the assumption that may be made in this case is that the defect candidate detected in the first image line 101 is a " false positive" result , and so the defect candidate is discarded as a false positive . As a result of this procedure , the defect detection method becomes more robust vis-a-vis noise that is superimposed on the piece of image information . The same applies accordingly to possible optical ef fects . The piece of information regarding the correctness then states that the photomask 17 is defect free in the region of the overlap, and this is output to the operator in text form by way of a screen . The advantage thereof is that the operator need not be occupied with assessing the incorrectly detected defect in the first image line 101 . In another example , a display coloured in yellow is used to indicate to the operator a piece of reliability information that signals that a defect was detected in at least one of the two image lines . Alternatively, a green colour, for example , may be displayed i f a defect was not detected in any of the image lines , representing a greater reliability . In this example , the operator in that case merely considers the defects indicated in yellow since these in particular require the attentiveness of a human evaluation .

[0060] Whether the defect candidate of the first piece of defect information corresponds to a defect candidate in the second piece of defect information is ascertained on the basis of a piece of identity information . In an example , the piece of identity information states that two defect candidates must be arranged within a certain radius on the photomask in order to be considered identical . The piece of identity information orthe radius was set by an operator of the mask inspection device before the method is performed.

[0061] In an embodiment, a first defect detection method optimized for the capture of an image line in the positive X-di- rection is used for the first image line 101. A second defect detection method optimized for the capture of an image line in the negative X-direction is used for the capture of the second image line 102. The influence of certain optical effects may be reduced thereby.

[0062] In a further embodiment, a third image line 103 is captured by virtue of the positioning system 26 displacing the photomask 17 in the positive X-direction. The third image line 103 also comprises a piece of image information regarding the region on the photomask 17 illuminated during the displacement. Before the third image line 103 is captured, the positioning system 26 displaces the photomask 17 in the positive Y-direction. In this case, the photomask 17 is displaced in the Y-direction in such a way that there is an overlap between the regions on the photomask 17 that in each case correspond to the first image line 101, the second image line 102 and the third image line 103.

[0063] Fig. 4 shows the first image line 101 (dashed line) , the second image line 102 (solid line) and the third image line 103 (dash-dotted line) in a corresponding manner on the photomask 17. There is a threefold overlap 300 between the three image lines 101, 102, 103, since said overlap results from the overlap of three image lines. The overlap parameter is approximately 0.75 in this example. On account of the threefold overlap 300, the first, second and third image lines 101, 102, 103 comprise a redundant piece of image information regarding the region on the photomask 17 that corresponds tothe overlap. In other words, the same piece of image information is contained once in the first image line 101, once in the second image line 102 and once in the third image line 103.

[0064] As is evident from Fig. 4, in addition to the threefold overlap 300 this example also comprises regions in which only two image lines overlap in a twofold overlap 200.

[0065] In an example, the defect detection method is also applied to the third image line 103 in order to determine a third piece of defect information. The first piece of defect information, the second piece of defect information and the third piece of defect information are compared with one another in order to determine the piece of information regarding the correctness of the structure of the photomask arranged within the overlap. The three pieces of defect information are thus combined with one another in order to obtain in-depth information regarding the correctness. In an example, the first piece of defect information comprises a defect candidate. If the second piece of defect information and the third piece of defect information also contain the defect candidate, i.e. if the defect detection method detects the same defect on the photomask 17 in all three image lines, then the defect candidate is verified as an actual defect. The actual defect is subsequently repaired in a repair step. If the second and / or third piece of defect information does not contain the defect candidate, then the defect candidate is discarded as a false positive in this example. In another example, a defect candidate is verified or discarded according to the majority principle. Thus, if at least two of the pieces of defect information contain the defect candidate, the latter is verified as an actual defect. If only one piece of defect information contains the defect candidate, then the latter is discarded as a false positive.

[0066] In a further example , the first image line 101 and the second image line 102 are superimposed on each other and combined by virtue of being averaged such that a combined image line is created . Since the piece of image information in the first and the second image line 101 , 102 is superimposed with uncontrollable noise , the combined image line has an improved SNR vis-a-vis the first image line 101 and the second image line 102 . The defect detection method is applied to the combined image line in order to determine a combined piece of defect information . In this example , the combined piece of defect information is compared with a further combined piece of defect information, which was also determined by the application of the defect detection method to an image line obtained from the combination of two image lines .

[0067] Should the piece of information regarding the correctness of the photomask 17 state that the photomask 17 comprises at least one defect , a repair step is performed in order to repair the defect on the photomask 17 . Should the piece of information regarding the correctness of the photomask 17 state that the photomask 17 does not comprise any defect , the defect- free photomask 17 is used in a method of microlithography in order to illuminate a substrate .

Claims

27Claims1. Method for inspecting a microlithographic photomask (17) , wherein with the aid of a mask inspection device comprising an illumination lens (16) and a projection lens (22) , the image of a photomask (17) illuminated by means of the illumination lens (16) is projected by the projection lens (22) onto an image sensor (24) of a camera (23) arranged in the image plane of the projection lens (22) , comprising the following steps: a) placing the photomask (17) on a positioning system (26) that is designed to displace the photomask (17) ; b) displacing the photomask (17) using the positioning system (26) such that the image sensor (24) captures a first image line (101) that corresponds to a first region on the photomask (17) ; c) displacing the photomask (17) using the positioning system (26) such that the image sensor (24) captures a second image line (102) that corresponds to a second region on the photomask (17) , with there being an overlap (200) between the first region and the second region such that the first and the second image line (101, 102) comprise a piece of image information regarding a structure of the photomask (17) arranged within the overlap (200) ; d) applying a defect detection method to the piece of image information in the first image line (101) in order to determine a first piece of defect information; e) applying a defect detection method to the piece of image information in the second image line (102) in order to determine a second piece of defect information;f) comparing the first piece of defect information with the second piece of defect information in order to determine a piece of information regarding the correctness of the structure of the photomask (17) arranged within the overlap (200) .

2. Method according to Claim 1, wherein the first image line(101) is captured by virtue of the positioning system (26) displacing the photomask (17) in a positive X-direction and wherein the second image line (102) is captured by virtue of the positioning system (26) displacing the photomask (17) in the positive or negative X-direction.

3. Method according to Claim 2, wherein the capture of the second image line is preceded by the positioning system (26) displacing the photomask (17) in a Y-direction such that an overlap parameter is less than 1.

4. Method according to Claim 3, wherein the overlap parameter is less than 0.1, preferably less than 0.05, further preferably less than 0.02 and further preferably less than 0.01.

5. Method according to any of the preceding claims, wherein the overlap (200) corresponds to a region on the photomask (17) that is specified on the basis of a specified region.

6. Method according to any of the preceding claims, wherein the determined piece of information regarding the correctness is compared with an expected correctness.

7. Method according to Claim 6, wherein the comparison between the determined piece of information regarding the correctness and the expected correctness is used to determine a piece of calibration information for the mask inspection device and / or a defect detection method, in order to reducea deviation between the determined piece of information regarding the correctness and the expected correctness.

8. Method according to any of the preceding claims, wherein the overlap (200) is specified on the basis of structures to be registered.

9. Method according to any of the preceding claims, wherein steps b) and c) are repeated in order to capture a plurality of pairs of first and second image lines (101, 102) and wherein steps d) to f) are performed on the basis of a specified rate for a subset of the plurality of pairs of first and second image lines (101, 102) .

10. Method according to any of the preceding claims, wherein a defect candidate in the first or second piece of defect information is verified by a defect candidate in the respective other piece of defect information such that the piece of information regarding the correctness of the structure of the photomask (17) arranged within the overlap (200) then states that an actual defect is present if a certain defect candidate is present in the first and in the second piece of defect information.

11. Method according to any of Claims 1 to 6, wherein the piece of information regarding the correctness of the structure of the photomask (17) arranged within the overlap (200) then states that an actual defect is present if a certain defect candidate is present in the first or in the second piece of defect information.

12. Method according to either of Claims 10 and 11, wherein the information regarding the correctness of the structure of the photomask (17) arranged within the overlap (200) comprises a piece of reliability information.

13. Method according to Claim 12, wherein the piece of reliability information specifies the number of pieces of defect information in which a certain defect candidate is present.

14. Method according to either of Claims 12 and 13, wherein a setting of the mask inspection device is modified in response to the piece of reliability information.

15. Method according to any of the preceding claims, wherein a specified piece of identity information is used to determine whether a defect candidate in the first piece of defect information corresponds to a defect candidate in the second piece of defect information.

16. Method according to Claim 15, wherein the piece of identity information states that two defect candidates are identical if they have the same centre or are arranged within a certain radius on the photomask (17) or have the same shape or if a shape difference lies within a certain shape tolerance .

17. Method according to any of the preceding claims, wherein the piece of information regarding the correctness of the structure of the photomask (17) arranged within the overlap (200) and / or the piece of reliability information is output, preferably acoustically, visually, as a database entry or as a text file.

18. Method according to any of the preceding claims, wherein at least one defect detection method comprises the application of a trained model from machine learning, with the input of the model comprising an image line (101, 102, 103) or a section of an image line and the output of the model comprising a piece of defect information.3119. Method according to any of the preceding claims, wherein the photomask (17) is displaced using the positioning system (26) such that the image sensor (24) captures a third image line (103) that corresponds to a third region on the photomask (17) , with there being an overlap (300) between the first region and the third region and also between the second region and the third region such that the first, the second and the third image line (101, 102, 103) comprise the image information regarding the structure of the photomask (17) arranged within the overlap (300) , and wherein the third piece of defect information is also compared with the first piece of defect information and the second piece of defect information in order to determine the piece of information regarding the correctness of the structure of the photomask (17) arranged within the overlap (300) .

20. Method according to Claim 19, wherein at least one image line is combined from two overlaid image lines.

21. Method according to any of the preceding claims, wherein the photomask (17) is an EUV or DUV mask.

22. Computer program product or a set of computer program products, comprising program parts which, when loaded into a computer or into networked computers, are designed to perform a method according to any of the preceding claims.

23. System for inspecting a microlithographic photomask (17) , comprising a mask inspection device and a computer configured to perform a method according to any of Claims 1 to 21.

24. Method for repairing a microlithographic photomask (17) , wherein a material of the photomask (17) is removed, added and / or structurally modified during a repair step,32 characterized in that the repair step is preceded by determining a piece of information regarding the correctness of the photomask (17) by means of a method according to any of Claims 1 to 21 and the repair step is subsequently performed in a region of the photomask (17) with an insufficient correctness.

25. Method according to Claim 24, wherein the piece of information regarding the correctness of the photomask (17) is determined by means of a method according to Claim 9, and the repair step is performed only if sufficient reliability is determined.

26. Method of microlithography, wherein with the aid of a projection exposure apparatus comprising an illumination lens(16) and a projection lens (22) , the image of a photomask(17) illuminated by means of the illumination lens (16) is projected by the projection lens (22) onto a substrate coated with a light-sensitive layer and arranged in the image plane of the projection lens (22) , characterized in that the exposure of the substrate is preceded by determining a piece of information regarding the correctness of the photomask (17) by means of a method according to any of Claims 1 to 21 and the exposure of the substrate is performed only if sufficient correctness and / or reliability is determined.

Images