METHOD AND APPARATUS FOR PROCESSING A LITHOGRAPHIC MASK

Abstract

Methods for repairing a defect of a lithographic mask with a particle beam are described. One such method can comprise the following steps: Processing the defect with the particle beam with a first set of processing parameters; processing the defect with the particle beam with a second set of processing parameters; wherein at least one parameter from the first set of processing parameters differs from the second set of processing parameters.

Claims

1. A method for repairing a defect of a lithographic mask with a particle beam, comprising: a. processing the defect with the particle beam with a first set of processing parameters; and b. processing the defect with the particle beam with a second set of processing parameters; c. wherein at least one parameter from the first set of processing parameters differs from the second set of processing parameters; d. wherein at least one marginal pixel of the defect is processed with the first set of processing parameters and at least one non-marginal pixel of the defect is processed with the second set of processing parameters.

2. The method of claim 1, wherein a first section of the defect is repaired with the aid of a first set of processing parameters and a second section of the defect is repaired with the aid of a second set of processing parameters, wherein the first section does not overlap the second section.

3. The method of claim 1, wherein the processing of a pixel of the defect is effected selectively either with the first or the second set of processing parameters, depending on the position of the pixel in the defect.

4. The method of claim 1, wherein at least one edge pixel of the defect is processed with the first set of processing parameters, and/or at least one non-edge pixel and/or volume pixel of the defect are/is processed with the second set of processing parameters.

5. The method of claim 1, wherein the at least one parameter from the first set of processing parameters comprises at least one of the following elements: a lower beam current, a lower beam flux, a lower beam fluence, a lower particle energy, or a smaller pixel size than defined by the second set of processing parameters.

6. An apparatus for repairing a defect of a lithographic mask with a particle beam, comprising: a. means for providing a particle beam; b. means for receiving information about the defect; and c. means for determining at least one first region of the defect for processing with the particle beam with a first set of processing parameters and at least one second region of the defect for processing with the particle beam with a second set of processing parameters, on the basis of the information; d. wherein at least one parameter from the first set of processing parameters differs from the second set of processing parameters; e. wherein the means for determining is configured such that the at least one first region has at least one marginal pixel of the defect and the at least one second region has at least one non-marginal pixel of the defect.

7. The apparatus of claim 6, wherein the information about the defect comprises information about the geometry of the defect.

8. A method for processing a lithographic mask, comprising: a. a first process, in which a particle beam with a first set of intrinsic beam parameters is guided onto the mask; and b. a second process, in which the particle beam with a second set of intrinsic beam parameters is guided onto the mask; c. wherein at least one parameter from the first set of intrinsic beam parameters differs from the second set of intrinsic beam parameters; d. wherein the at least one parameter from the first set of intrinsic beam parameters comprises a higher particle energy than defined by the second set of intrinsic beam parameters.

9. The method of claim 8, wherein the first process is a diagnostic process, in particular a process for correcting a drift, and/or the second process is a repair process.

10. The method of claim 8, wherein the first set of intrinsic beam parameters is configured such that the secondary electron contribution lies in the range of 0.5 to 1.5, preferably in the range of 0.75 to 1.25, and/or wherein the second set of intrinsic beam parameters is configured such that the secondary electron contribution is greater than 1.5, preferably greater than 1.75.

11. The method of claim 1, furthermore comprising providing at least one precursor gas, such that the particle beam excites a chemical reaction of the at least one precursor gas at the mask.

12. An apparatus or processing a lithographic mask, comprising: a. means for guiding a particle beam onto the mask; b. means for receiving information about a first and a second process; and c. means for determining a first and a second set of intrinsic beam parameters for the first and the second process, respectively, with which the particle beam is intended to be directed onto the mask; d. wherein at least one parameter from the first set of intrinsic beam parameters differs from the second set of intrinsic beam parameters; e. wherein the at least one parameter from the first set of intrinsic beam parameters comprises a higher particle energy than defined by the second set of intrinsic beam parameters.

13. The apparatus of claim 12, wherein the information identifies the first process as a diagnostic process and/or the second process as a repair process.

14. The apparatus of claim 6, furthermore comprising means for providing at least one precursor gas, such that the particle beam excites a chemical reaction of the at least one precursor gas at the mask.

15. A computer program comprising executable instructions that are designed, when executed by a computer, to carry out the steps of the method of claim 1.

16. A computer program comprising executable instructions that are designed, when executed by a computer, to carry out the steps of the method of claim 8.

17. The method of claim 2, wherein the processing of a pixel of the defect is effected selectively either with the first or the second set of processing parameters, depending on the position of the pixel in the defect.

18. The method of claim 2, wherein at least one edge pixel of the defect is processed with the first set of processing parameters, and/or at least one non-edge pixel and/or volume pixel of the defect are/is processed with the second set of processing parameters.

19. The method of claim 2, wherein the at least one parameter from the first set of processing parameters comprises at least one of the following elements: a lower beam current, a lower beam flux, a lower beam fluence, a lower particle energy, or a smaller pixel size than defined by the second set of processing parameters.

20. The method of claim 9, wherein the first set of intrinsic beam parameters is configured such that the secondary electron contribution lies in the range of 0.5 to 1.5, preferably in the range of 0.75 to 1.25, and/or wherein the second set of intrinsic beam parameters is configured such that the secondary electron contribution is greater than 1.5, preferably greater than 1.75.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0074] The following detailed description describes currently preferred exemplary embodiments of the invention, with reference being made to the following drawings:

[0075] FIGS. 1A-1C: Example of a defect of a photomask and a classification of the defect into a set of marginal pixels and a set of non-marginal pixels and also into a set of edge pixels and non-edge pixels.

[0076] FIGS. 2A-2B: Cross-sectional view for interaction volumes of the particle beam and secondary electron contribution at high and respectively low particle energy.

[0077] FIG. 3: Schematic diagram of the secondary electron contribution as a function of the particle energy.

[0078] FIG. 4: Flow diagram for a method for repairing a defect of a lithographic mask with a particle beam.

[0079] FIG. 5: Schematic diagram for an apparatus for repairing a defect of a lithographic mask with a particle beam.

DETAILED DESCRIPTION

[0080] Currently preferred embodiments of the present invention are explained in greater detail below with reference to the drawings.

[0081] FIGS. 1A-1C show an exemplary lithographic mask 100 (hereinafter for short: mask 100) with a defect 120 in accordance with one example.

[0082] FIG. 1A shows an exemplary pattern 110 of the mask 100. The pattern 110 is illustrated as an absorbent pattern comprising horizontally illustrated strips of absorbent material that absorbs radiation during exposure. Radiation that impinges on the photomask 100 between the strips during exposure can penetrate through the photomask and expose the object, e.g., a wafer coated with photosensitive material, on the opposite side of the photomask. One of the strips has a defect 120. The absorbent material is at least partly missing in the region of the defect 120. The defect 120 is illustrated by way of example as a rectangular structure. During exposure, radiation that impinges on the region of the defect 120 can therefore penetrate by mistake at least partly through the mask 100. The defect 120 is thus illustrated by way of example as a clear defect 120. In order to repair the mask 100, absorbent material can be deposited in the region of the defect 120. The geometry of the pattern 110 and also the nature of the defect as a clear defect are merely by way of example.

[0083] Besides two-dimensional coordinates (e.g., x, y coordinates) that can extend in the mask plane, the defect 120 can also have three-dimensional coordinates (e.g., z coordinate) that can indicate a depth of the defect.

[0084] In the prior art, defects such as the defect 120 were repaired with exactly the same beam and process parameters. This is not always optimal, particularly in the case of large-area defects.

[0085] In order to improve the repair, the defect 120, as shown in FIGS. 1B and 1C, is divided into a first region 121 and a second region 122. The first and respectively the second region can be formed by a first and respectively second (geometrically) continuous set of pixels of the defect 120 (or of the repair shape of the defect 120).

[0086] The first set of pixels (region 121) can be formed for example by one or more non-marginal pixels of the defect 120. The second set of pixels (region 122) can be formed for example by one or more marginal pixels of the defect 120.

[0087] The first set of pixels 121 can be exposed for example with beam and/or process parameters that lead to a processing mode with relatively low resolution, relatively low accuracy, but relatively high throughput. For example, a high beam current, a high particle energy, a high gas flow and/or a large pixel size (e.g., of the particle beam) can be used.

[0088] The second set of pixels 122 can be exposed for example with beam and/or process parameters that lead to a processing mode with relatively high resolution, relatively high accuracy, but relatively low throughput. For example, a low beam current, a low particle energy, a low gas flow and/or a small pixel size (e.g., of the particle beam) can be used.

[0089] Alternatively or additionally, the second set of pixels 122 (marginal pixels) can be subdivided into edge pixels 122a and non-edge pixels 122b. The edge pixels 122a can be exposed for example with beam and/or process parameters that lead to a processing mode with relatively high resolution, relatively high accuracy, but relatively low throughput. For example, a low beam current, a low particle energy, a low gas flow and/or a smaller pixel size (e.g., of the particle beam) can be used in comparison with beam and/or process parameters for the non-marginal pixels 121 and/or for the non-edge pixels 122b. The same beam and/or process parameters as for the non-marginal pixels 121 can be used for the non-edge pixels 122b. However, it is also possible for a lower beam current, a lower particle energy, a lower gas flow and/or a smaller pixel size (e.g., of the particle beam) to be used for the non-edge pixels 122b in comparison with beam and/or process parameters for the non-marginal pixels 121.

[0090] The processing of the first set of pixels 121 and of the second set of pixels 122 and also of the sets 122a and 122b can be carried out simultaneously, in an alternating manner (in parallel) or temporally successively.

[0091] For example, for the marginal pixels and/or edge pixels, a pixel size of the of the repair shape of 1-10 nm.sup.2, preferably 1-5 nm.sup.2, can be provided, e.g., 1*1 nm.sup.2, 2*2 nm.sup.2. For the non-marginal pixels and/or non-edge pixels and/or volume pixels, a pixel size of the repair shape can be larger, e.g., 5-25 nm.sup.2, preferably 10-25 nm.sup.2, e.g., 2*2 nm.sup.2, 3*3 nm.sup.2, 4*4 nm.sup.2 or 5*5 nm.sup.2.

[0092] For example, for the marginal pixels and/or edge pixels, a pixel size of the particle beam of 1-10 nm.sup.2, preferably 1-5 nm.sup.2, can be provided, e.g., 1*1 nm.sup.2, 2*2 nm.sup.2. For the non-marginal pixels and/or non-edge pixels and/or volume pixels, a pixel size of the particle beam can be larger, e.g., 5-40 nm, preferably 10-25 nm.sup.2, e.g. 2*2 nm.sup.2, 3*3 nm.sup.2, 4*4 nm.sup.2, 5*5 nm.sup.2 or 6*6 nm.sup.2.

[0093] It is also possible to indicate the pixel size (of the particle beam or of the repair shape) in units of the minimum focus diameter of the particle beam.

[0094] The assignment of the pixels of a repair shape for a defect 120 to one or more sets of pixels can also be regarded as segmentation of the repair shape into one or more sub-repair shapes (each of which comprises a set of pixels). The processing of the sets of pixels can then be effected in particular in temporal succession, namely by sequential processing of the individual sub-repair shapes. The sequence of repair shapes and the respective process steps and parameters thereof can be stored in a common file, for example. The latter can then be processed by a corresponding apparatus, as described herein.

[0095] FIGS. 2A-2B show the effects of a change in the particle energy during the processing of a substrate 200 of a mask by way of example on the basis of an electron beam 220. The electron beam 220 can be incident on the substrate 200, e.g., approximately perpendicularly. A relatively high particle energy of the electrons leads to a relatively large interaction volume 225 with the substrate 220 (FIG. 2A). A relatively low particle energy leads to a relatively small interaction volume with the substrate (FIG. 2B). The interaction volume 225 is approximately drop-shaped in cross section, with the tip of the drop oriented in the direction of the incident beam 220, whereas the drop can have an approximately round shape on the side which is opposite to the incident beam 220. The depth of the drop generally increases with the particle energy, just like the average diameter of the drop in the direction of the particle beam.

[0096] By use of a suitable choice of the particle energy (e.g., by use of adapting the accelerating voltage), it is thus possible to set the “coarseness” of the grid with the aid of which the repair is effected—both in the plane of the mask and in the z-direction. By way of example, simulations of the secondary electrons from a quartz surface allow the following estimation: At an accelerating voltage of 1 kV, secondary electrons emerge in a radius of up to ˜25 nm around the impingement location of the primary beam. At an accelerating voltage of only 0.6 kV or 0.2 kV, said radius decreases so as to be up to ˜15 nm or up to ˜10 nm, respectively.

[0097] Situated around the tip of the drop there are generally zones 240 from which secondary electrons are released upon bombardment with the particle beam. The size or extent of these zones 240 of the substrate also depends on the particle energy of the incident particle beam. At a high particle energy said zones tend to be somewhat deeper and also to have a larger surface area in the substrate plane. Exit surfaces for secondary electrons thus form at the top side facing the particle beam of the substrate.

[0098] FIG. 3 shows an exemplary functional relationship 300 between the particle energy and the secondary electron yield. The relationship is shown by way of example for an electron beam and the yield of secondary electrons n per incident electron is shown as a function of the energy of the electrons (in keV). With higher energy of the electrons, firstly the yield of secondary electrons rises from a value close or equal to zero. At a particle energy of approximately 0.15 keV, the secondary electron yield is then approximately n=1. This point is identified by the reference sign 340 in FIG. 3. Subsequently, the rise of n levels out somewhat and reaches a local maximum 320 of approximately n=2 at a particle energy of approximately 0.5 keV. Subsequently, the yield decreases again, reaches a value of n=1 (reference sign 360) again at an energy of approximately 2 keV and then decreases further with higher particle energy.

[0099] In some exemplary embodiments, it may be helpful, for the purpose of processing a lithographic mask, to carry out a first process with a different set of intrinsic beam parameters in comparison with a second process. First and respectively second processes in this sense can be, e.g.: Image recording processes for generating a repair shape, process for creating at least one drift correction marker, actual repair process (e.g., particle-beam-induced etching and/or deposition), process for drift correction during the actual repair process, end pointing process (which ascertains whether material has already been removed to the desired depth and/or whether the material has already been deposited with the desired thickness), image recording process for verifying the repair.

[0100] In some exemplary embodiments, it may be helpful, for the purpose of processing a lithographic mask, to carry out a first process with a different accelerating voltage for the particle beam (e.g., electrons or ions) in comparison with a second process. For example, a process for generating a repair shape and/or a process for imaging can be carried out with a different accelerating voltage in comparison with the actual repair process on the basis of the generated repair shape. The accelerating voltages can each be optimized independently of one another even though the processes serve for repair and diagnostics of one and the same defect and can be carried out without removing the mask from the apparatus that provides the particle beam. In accordance with one example, the voltages can each be optimized for different successive processes (e.g., image recordings and/or repair steps) concerning the same defect.

[0101] In accordance with one example, an imaging process (for creating the repair shape and/or for checking the repair that has been effected) is carried out with a higher accelerating voltage in order to obtain a high image resolution. For this purpose, for example, backscattered electrons/ions can be detected, e.g., with the aid of the technique of energy selective backscattering (EsB). Conversely, for the subsequent repair process (with local etching and/or local deposition with the aid of a precursor gas), a lower accelerating voltage can be used in order to obtain a high repair resolution. This is because with a lower accelerating voltage the extent of the zones 240 (cf. FIG. 2) from which secondary electrons are released upon bombardment with the particle beam is smaller. Accordingly, the local chemical reaction used for the repair is locally better delimited and the repair resolution is increased. By way of example, for the actual repair in the case of an electron beam and a mask for EUV, it is possible to use an accelerating voltage of 50 to 1200 V, of 100 to 1200 V, of 200 to 1200 V, of 300 to 1000 V, of 400 to 800 V, or of 500 to 700 V, preferably 550 to 650 V or approximately 600 V. For the imaging process, it is possible to use, e.g., a voltage of >2 kV, for example 2 to 5 kV, 2 to 4 kV, 2,5 to 3,5 kV or approximately 3 kV.

[0102] In a further example, the accelerating voltage for the imaging process can be chosen such that the secondary electron yield is approximately 1. This may be helpful in particular if sample charging effects might result in problems. In the case of an electron beam, the accelerating voltage can have values of approximately 300 to 700 V, 400 to 600 V, 450 to 550 V, or approximately 500 V. For the actual repair, it is possible to use by example (e.g., in the case of an electron beam and/or a mask for EUV) an accelerating voltage which allows a desired resolution, e.g., an optimum resolution using 50 to 1200 V, 100 to 1200 V, 200 to 1200 V, 300 to 1000 V, 400 to 800 V, or 500 to 700 V, preferably 550 to 650 V or approximately 600 V.

[0103] In a further example, the accelerating voltage for the imaging process (before or after or else during the actual repair process) can be chosen such that a material contrast is optimized (e.g., by energy selective detection of backscattered electrons by use of an EsB detector). In another example, the accelerating voltage for the imaging process (before or after or else during the actual repair process) can be chosen such that edge effects are minimized (edges appear at the correct position). Moreover, it is possible to choose the accelerating voltage for the imaging process (before or after or else during the actual repair process) such that edge effects are maximized (edges are then more clearly recognizable). For the actual repair, in each case by way of example (e.g., in the case of an electron beam and/or an EUV mask), it is possible to use a different accelerating voltage which allows a desired repair resolution, e.g., an optimum resolution using 50 to 1200 V, 100 to 1200 V, 200 to 1200 V, 300 to 1000 V, 400 to 800 V, or 500 to 700 V, preferably 550 to 650 V or approximately 600 V.

[0104] In the above examples, the accelerating voltage is merely by way of example. Other voltages (e.g., sample bias voltage) can also be varied and accordingly used as various parameters for first and second processes.

[0105] In some examples, for the purpose of processing a lithographic mask, different pixel sizes are used for a first and a second process. The pixel sizes can be optimized independently of one another for the respective process.

[0106] In accordance with one example, the repair shape is generated with a first pixel size. The actual repair can be effected (at least partly) with a different pixel size (of the particle beam). By way of example, the pixels of the repair shape can be divided into a first and respectively a second set of pixels. The pixel size of the particle beam for repairing the first and second sets can then be optimized in each case for the first and respectively the second set, as described herein.

[0107] However, it is also possible to use a different pixel size for the actual repair without dividing the repair shape into a plurality of sets of pixels. For example, the pixel size (of the particle beam) can be chosen depending on the geometry of the repair shape. It can be chosen such that the repair can be effected with as far as possible an integral number of pixels. It can be chosen depending on a total area of the defect and/or a complexity of a contour of the defect.

[0108] It is also possible for at least one further parameter (beam and/or process parameter) to be optimized toward the pixel size used in the respective process. In the apparatuses of the present invention, for example, an automatic optimization of the further parameters toward the pixel size respectively used can be effected which optionally includes an (automatic) check as to whether the pixel size respectively selected is expedient and/or allowed. It is thus possible to prevent the (isolated) selection of a specific pixel size from resulting in errors in the process.

[0109] In accordance with one example, two repair processes are carried out. In a first process, an overcorrection is effected. For example, material is deposited over a larger area (or etched away over a larger area) than that corresponding to the extent of the defect. In a second step, the deposited material is then partly etched away again (or, respectively, material is deposited), such that the remaining material reproduces the contours of the defect as well as possible. For example, the first process can be carried out with one or more parameters for low resolution. For example, the second process can be carried out with one or more parameters for high resolution. This can be advantageous in order to rapidly realize a (large-area) repair approximately corresponding to the defect. In a second step, the refinements necessary for the precise repair can then be carried out. The (slow) process with high resolution then need only be carried out across a smaller area. In another example, the first process can be carried out with one or more parameters for high resolution and the second process can be carried out with one or more parameters for low resolution.

[0110] In particular as a result of varying the beam and/or process parameters, it may be necessary to ensure matching between different images. This is because the mask may expand or warp differently on account of the various possible thermal loads. Moreover—in particular when using an electron beam—the particle beam may be deflected to different extents as a result of different static charging. For this purpose, it is possible to create and/or use markings on the mask. However, it is also possible to ensure matching by use of a prediction on the basis of a model. One example of this is disclosed in DE 10 2018 209 562 B3. Further development of the model for different parameters—as described in the present case—may make it unnecessary to use markings on the mask. Especially since markings also degrade over time and should therefore be scanned as infrequently as possible, this is advantageous in particular if—as in the present case—different parameters are intended to be used and the need for an adaptation of the correction of displacements or distortions thus increases. By way of example, it is possible to ascertain empirically what changes in specific parameters cause what changes in the images. However, a model that physically calculates the changes and/or predicts them on the basis of machine learning is also possible.

[0111] An apparatus for processing a lithographic mask and/or for repairing a defect of a lithographic mask can comprise the following: (a) Means for recording measurement data while the mask is exposed to the particle beam; and (b) means for predetermining a drift of the beam of charged particles relative to the lithographic mask with a trained machine learning model and/or a predictive filter, wherein the trained machine learning model and/or the predictive filter use(s) at least the measurement data as input data. In particular, the predictive filter and/or the trained model can make it possible to ascertain from a drift for a specific set of processing parameters (beam parameters and/or process parameters) a drift for a changed set of processing parameters, such that the correction of the drift upon a change of the processing parameters in a second process only has to be effected relative to the correction already effected in a first process.

[0112] A further aspect to be considered is that the Point Spread Function (PSF) of the imaging system can be adapted, if appropriate, taking account of material effects of the mask, particularly if parameters are changed in the processes—as described in the present case. For this purpose, provision can be made for measuring the PSF. The measurement can be effected for different sets of parameters that are also used subsequently. However, it is also possible for one or more measurements to be effected only for one or more selected sets of parameters, from which the PSF is then ascertained for the sets of parameters used subsequently in each case. For this purpose, it is possible to effect an interpolation or extrapolation or a complex conversion on the basis of mathematical models.

[0113] The measurement can be effected for example with the aid of a sample as described for instance in DE 10 2018 210 522 A1. Alternatively or additionally, two or more images of a reference structure can be recorded specifically under different recording conditions. PSFs can be generated for the two or more recorded images with a modified reference image of the reference structure, for example by recording images in the form of a focus stack.

[0114] The image recorded in each case can then be deconvolved with the PSF respectively measured or ascertained on the basis of measurement data, optionally taking account of material effects. Finally, the repair shape can be generated from the image thus sharpened.

[0115] FIG. 4 shows an exemplary flow diagram for a method for processing a lithographic mask or respectively for repairing a defect of the mask.

[0116] The method can firstly comprise a process 440 for creating drift correction markers. This optional process 410 can be carried out at the start of the method. However, it is also possible to carry it out at a later point in time. The drift correction markers can be observed—e.g., during processing—and so a drift or a distortion can be corrected.

[0117] The method can additionally comprise the process of recording 410 image information with regard to a defect. The recording can be effected using a particle beam, for example by use of an electron beam, which can be provided by an SEM (cf. FIG. 5).

[0118] Creating a repair shape 420 can be provided as a further process. The repair shape can be created for example by a computer system (cf. FIG. 5). For this purpose, information from the recording in accordance with process 410 can be used, for example. The creating can be effected by a computer system integrated into an apparatus for providing the particle beam (for example into an apparatus comprising an SEM, cf. FIG. 5). Alternatively, however, the computer system can also obtain the corresponding information from the recording by way of an interface, for example by way of a user interface or some other interface (to a data carrier, server, Internet, etc.). The computer system can also obtain information from some other, e.g., previously used, measuring apparatus (e.g., AIMS, AFM (Atomic Force Microscope)) and use it to create the repair shape. Furthermore, inputs by the user can also be taken into account in the process of creating the repair shape by use of the computer system, which inputs can be received by way of a user interface.

[0119] In a further step 430, at least one first pixel of the defect and at least one second pixel of the defect can be selected. The selection can be effected at least partly manually by the user with the aid of the user interface. However, the selection can also be effected at least partly automatically by the computer system.

[0120] In a further step (not illustrated), it is possible to define a first and a second set of processing parameters for the processing of the at least one first pixel and respectively the at least one second pixel with the aid of the computer system. This defining can also be effected at least partly automatically and/or at least partly manually with the aid of the user interface.

[0121] In a further process 450, the defect, in particular the at least one first pixel, can be processed with the particle beam and the first set of processing parameters. The further process 460 of drift correction can be effected during this processing.

[0122] In an additional process 470, the defect, in particular the at least one second pixel, can be processed with the particle beam and the second set of processing parameters. During this processing, too, the further process 460 of drift correction can be effected, but usually with changed correction parameters.

[0123] Finally, it is possible to effect a further process of recording image information with regard to the defect for “end pointing” 480 and/or recording image information for verifying the repair of the defect 490.

[0124] FIG. 5 shows a schematic section through some important components of an apparatus 500 which can be used for processing and repairing (one or more defects of) a lithographic mask. The exemplary apparatus 500 in FIG. 5 comprises a modified scanning particle microscope 510 in the form of a scanning electron microscope (SEM) 510.

[0125] The apparatus 500 comprises a particle beam source 505 in the form of an electron beam source 505, which generates an electron beam 515 as particle beam 515. An electron beam 515 has the advantage—compared with an ion beam—at the electrons that impinge on the sample 525 or the lithographic mask substantially cannot damage the sample or the mask. However, it is also possible to use an ion beam, an atomic beam, a molecular beam or a photon beam (not illustrated in FIG. 5) for the purpose of processing the sample 525 in the apparatus 500.

[0126] The scanning particle microscope 510 is composed of an electron beam source 505 and a column 520, in which is arranged the beam optical unit 513 for instance in the form of an electron optical unit of the SEM 510. In the SEM 510 in FIG. 5, the electron beam source 505 generates an electron beam 515, which is directed as a focused electron beam 515 onto the sample 525, which can comprise the lithographic mask, at the location 522 by the imaging elements arranged in the column 520, said imaging elements not being illustrated in FIG. 5. The beam optical unit 513 thus forms the imaging system 513 of the electron beam source 505 of the SEM 500.

[0127] Further, the imaging elements of the column 520 of the SEM 510 can scan the electron beam 515 over the sample 525. The sample 525 can be examined using the electron beam 515 of the SEM 510.

[0128] The backscattered electrons and secondary electrons generated by the electron beam 515 in the interaction region of the sample 525 are registered by the detector 517.

[0129] The detector 517 arranged in the electron column 520 is referred to as an “in lens detector.” The detector 517 can be installed in the column 520 in various embodiments. The detector 517 converts the secondary electrons generated by the electron beam 515 at the measurement point 522 and/or the electrons backscattered from the sample 525 into an electrical measurement signal and transmits the latter to an evaluation unit 585 of a computer system 580 of the apparatus 500. The detector 517 can contain a filter or a filter system in order to discriminate the electrons in terms of energy and/or solid angle (not reproduced in FIG. 5). The detector 517 is controlled by a setting unit 590 of the apparatus 500.

[0130] The exemplary apparatus 500 can include a second detector 519. The second detector 519 can be designed to detect electromagnetic radiation, in particular in the X-ray range. As a result, the detector 519 makes it possible to analyze a material composition of the radiation generated by the sample 525 during the examination thereof. The detector 519 is likewise controlled by the setting unit 590.

[0131] Further, the apparatus 500 can comprise a third detector (not illustrated in FIG. 5). The third detector is often embodied in the form of an Everhart-Thornley detector and typically arranged outside the column 520. As a rule, it is used to detect secondary electrons.

[0132] The apparatus 500 can comprise an ion source that provides ions with low kinetic energy in the region of the sample 525 (not illustrated in FIG. 5). The ions with low kinetic energy can compensate for charging of the sample 525. Furthermore, the apparatus 500 can comprise a mesh at the output of the column 520 of the modified SEM 510 (likewise not shown in FIG. 5). Electrostatic charging of a sample 525 can likewise be compensated for by applying a voltage to the mesh. It is furthermore possible to earth the mesh.

[0133] The sample 525 is arranged on a sample stage 530 or a sample holder 530 for examination purposes. A sample stage 530 is also known as a “stage” in the art. As symbolized by the arrows in FIG. 5, the sample stage 530 can be moved in three spatial directions relative to the electron beam 515 of the SEM 510, for example by way of micro-manipulators that are not illustrated in FIG. 5.

[0134] Besides the translational movement, the sample stage 530 can be rotated at least about an axis oriented parallel to the beam direction of the particle beam source 505. It is furthermore possible for the sample stage 530 to be embodied such that it is rotatable about one or two further axes, this axis or these axes being arranged in the plane of the sample stage 530. The two or three axes of rotation preferably form a rectangular coordinate system. As can be gathered from FIG. 5, the rotation of the sample stage 530 about an axis of rotation that is arranged in the plane of the sample stage 530 is often possible only to a limited extent on account of the small distance between the end of the column and the sample 525.

[0135] The sample 525 to be examined can be any arbitrary microstructured component or device that requires analysis and, if appropriate, subsequent processing, for example the repair of a local defect of a lithographic mask. In this regard, the sample 525 can comprise for example a transmissive or a reflective photomask and/or a template for nanoimprint technology. The transmissive and the reflective photomask can comprise all types of photomasks, for instance binary masks, phase-shifting masks, OMOG (opaque MoSi on glass) masks, or masks for a double or multiple exposure.

[0136] Further, the apparatus 500 in FIG. 5 can comprise one or more scanning probe microscopes, for example in the form of an atomic force microscope (AFM) (not shown in FIG. 5), which can be used to analyze and/or process the sample 525.

[0137] The scanning electron microscope 510 illustrated by way of example in FIG. 5 is operated in a vacuum chamber 570. In order to generate and maintain a reduced pressure required in the vacuum chamber 570, the SEM 510 in FIG. 5 has a pump system 572.

[0138] The apparatus 500 includes a computer system 580. The computer system 580 can form the means described herein for receiving information about a first and a second process and/or the means for determining a first and a second set of intrinsic beam parameters for the first and respectively the second process.

[0139] Moreover, the computer system 580 can form the means described herein for receiving information about the defect and/or the means for determining at least one first pixel of the defect for processing with the particle beam with a first set of processing parameters and at least one second pixel of the defect for processing with the particle beam with a second set of processing parameters, on the basis of the information.

[0140] The computer system 580 can also comprise a scanning unit 582, which scans the electron beam 515 over the sample 525. Furthermore, it can comprise a setting unit 590 in order to set the various parameters of the modified scanning particle microscope 510 of the apparatus 500. Furthermore, the setting unit 590 can control the deflection system and rotation of the sample stage 530.

[0141] Moreover, the computer system 580 can comprise an evaluation unit 585, which analyzes the measurement signals of the detectors 517 and 519 and generates an image of the sample 525 therefrom, which image can be displayed in a graphical user interface of the computer system 580, which can comprise a display 595 and input means. In particular, the evaluation unit 585 can be designed to determine the position and a contour of a defect of missing material and/or a defect of excess material of a sample 525, for instance of the lithographic mask, from the measurement data of the detector 517. The evaluation unit 585 can additionally contain one or more algorithms which make it possible to determine a repair shape corresponding to the analyzed defect of the mask. The evaluation unit 585 of the computer system 580 can additionally include one or more algorithms that can ascertain the parameters (beam and/or process parameters) of specific processes. The algorithms of the evaluation unit 585 can be implemented using hardware, software or a combination thereof. In particular, the algorithm(s) can be realized in the form of an ASIC (Application Specific Integrated Circuit) and/or an FPGA (Field Programmable Gate Array).

[0142] The computer system 580 and/or the evaluation unit 585 can include a memory (not illustrated in FIG. 5), preferably a non-volatile memory, which stores one or more models of repair shapes for various mask types. The evaluation unit 585 can be designed to calculate, on the basis of a repair model, a repair shape for the defect(s) of the lithographic mask from the measurement data of the detector 517. Furthermore, the computer system 580 can comprise an interface 587 for exchanging data with the Internet, an Intranet and/or some other apparatus. The interface 587 can comprise a wireless or a wired interface.

[0143] As indicated in FIG. 5, the evaluation unit 585 and/or the setting unit 590 can be integrated into the computer system 580. However, it is also possible to embody the evaluation unit 585 and/or the setting unit 590 as independent unit(s) within or outside the apparatus 500. In particular, the evaluation unit 585 and/or the setting unit 590 can be designed to carry out some of their tasks by use of a dedicated hardware implementation.

[0144] The computer system 580 can be integrated into the apparatus 500 or can be embodied as an independent device (not shown in FIG. 5). The computer system 580 can be embodied using hardware, software, firmware or a combination.