Illumination optical unit for EUV projection lithography, and optical system comprising such an illumination optical unit

Abstract

An illumination optical unit for EUV projection lithography guides illumination light to an illumination field, in which a lithography mask can be arranged. A facet mirror with a plurality of facets guides the illumination light to the illumination field. Respectively one illumination channel which guides an illumination light partial beam is predetermined by one of the facets. Exactly one illumination channel is guided over respectively one of the facets. The illumination optical unit is configured so that, at any time and at any point in the illumination field when the illumination optical unit is in operation, any pairs of illumination light partial beams guided over different illumination channels are incident on this illumination field point at times of incidence, the time difference of which is greater than a coherence duration of the illumination light.

Claims

1. An illumination optical unit configured to guide illumination light to an illumination field, the illumination optical unit comprising: a first facet mirror comprising a plurality of first facets configured to guide the illumination light to the illumination field, wherein during operation of the illumination optical unit: the illumination light comprises a plurality of illumination light partial beams; each illumination light partial beam is guided by a respective illumination channel; each illumination channel is guided by a respective first facet; no more than one illumination channel is guided by a given first facet; the plurality of illumination light partial beams comprises a first illumination light partial beam and a second illumination light partial beam which is different from the first illumination light partial beam; the first illumination light partial beam is guided to the illumination field by a first illumination light channel; the second illumination light partial beam is guided to the illumination field by a second illumination light channel which is different from the first illumination light channel; the first and second illumination light partial beams are incident simultaneously at the same point in the illumination field; the first illumination light partial beam has a first travel time to the illumination field; the second illumination light partial beam has a second travel time to the illumination field; a difference between the first and second travel times is greater than a coherence duration of the illumination light; and the illumination optical unit is an EUV illumination optical unit.

2. The illumination optical unit of claim 1, further comprising a second facet mirror downstream of the first facet mirror in a beam path of the illumination light through the illumination optical unit, the second facet mirror comprising a plurality of second facets, wherein during operation of the illumination optical unit: each illumination channel is guided by a respective second facet; no more than one illumination channel is guided by a given second facet.

3. The illumination optical unit of claim 2, wherein: the first facet mirror is a field facet mirror; the first facets are field facets; the second facet mirror is as pupil facet mirror; the second facets are pupil facets; the illumination optical unit comprises a transmission optical unit configured to superimpose imaging of the field facets in the illumination field; and the transmission optical unit comprises the pupil facet mirror; the illumination optical unit is arrangeable so that images of the illumination light source come to rest at the location of the pupil facets; the pupil facet mirror is configured with a tilt so that the various illumination channels, by which each one of the field points of the illumination field is impinged upon by the illumination light partial beams during operation of the illumination optical unit, respectively have different illumination channel lengths between the light source and respectively one of the field points of the illumination field.

4. The illumination optical unit of claim 2, wherein: the first facet mirror is a field facet mirror; the first facets are field facets; the second facet mirror is embodied as pupil facet mirror; the second facets are pupil facets; the illumination optical unit comprises a transmission optical unit configured to superimpose imaging of the field facets in the illumination field; the transmission optical unit comprises the pupil facet mirror; the illumination optical unit is arrangeable so that images of the illumination light source come to rest at the location of the pupil facets; a main mirror surface of the pupil facet mirror deviates from a plane reference surface so that the various illumination channels, by which each one of the field points of the illumination field are impinged upon by the illumination light partial during operation of the illumination optical unit, respectively have different illumination channel lengths between the light source and respectively one of the field points of the illumination field.

5. The illumination optical unit of claim 2, wherein the illumination optical unit is configured so that during operation of the illumination optical unit: travel time differences exist between the illumination light partial beams depending on differences in path lengths of beam paths of the illumination channels; and between a location in the beam path of the illumination light prior to the division of the illumination light into the illumination light partial beams and the illumination field, each difference between the travel times of two different illumination light partial beams is always greater than a coherence duration of the illumination light for each location in the illumination field.

6. The illumination optical unit of claim 2, further comprising an optical retardation component configured to divide at least one illumination light partial beam into a plurality of partial beam components, wherein: between a first location in the beam path of the illumination light and the illumination field, the partial beam components have such pairwise travel time differences amongst themselves that each difference between the travel times of the partial beam components of any illumination light partial beams is greater than the coherence duration of the illumination light; and the first location lies both upstream of the division of the illumination light into the illumination light partial beams and upstream of the retardation component the partial beam components.

7. The illumination optical unit of claim 2, further comprising a scanning device configured to scan an illumination beam over the first facets so that that the illumination light illuminates the illumination field during operation of the illumination optical unit, wherein: when incident on the first facet mirror, an entire beam crosssection of the illumination beam is greater than 20% of a reflection surface of one of the first facets; and the scanning device is configured so that at any given time a given field point in the illumination field is illuminated via at most one of the first facets.

8. The illumination optical unit of claim 7, wherein the scanning device is configured so that a scanning region extending over a plurality of first facets is scanned on the facet mirror during operation of the illumination optical unit.

9. The illumination optical unit of claim 2, further comprising a scanning device configured to scan an illumination beam over the first facets so that the illumination light illuminates the illumination field during operation of the illumination optical unit, wherein: when incident on the first facet mirror, an entire beam crosssection of the illumination beam is greater than 200% of a reflection surface of one of the first facets; and the scanning device is configured so that at any given time a given field point in the illumination field is illuminated only via first facets in which the optical path length difference of their corresponding illumination chancels differs by at least one coherence length.

10. The illumination optical unit of claim 2, wherein the illumination optical unit is configured so that during operation of the illumination optical unit: illumination channel pairs exist in which illumination light guided along the illumination channel pairs is incident at at least one point of the illumination field so that the time difference is less than a coherence duration of the illumination light; and a cross-sectional area of at least one of the illumination channels of the illumination channel pairs does not contribute to illuminating the illumination field.

11. The illumination optical unit of claim 1, wherein the illumination optical unit is configured so that during operation of the illumination optical unit: travel time differences exist between the illumination light partial beams depending on differences in path lengths of beam paths of the illumination channels; and between a location in the beam path of the illumination light prior to the division of the illumination light into the illumination light partial beams and the illumination field, each difference between the travel times of two different illumination light partial beams is always greater than a coherence duration of the illumination light for each location in the illumination field.

12. The illumination optical unit of claim 1, further comprising an optical retardation component configured to divide at least one illumination light partial beam into a plurality of partial beam components, wherein: between a first location in the beam path of the illumination light and the illumination field, the partial beam components have such pairwise travel time differences amongst themselves that each difference between the travel times of the partial beam components of any illumination light partial beams is greater than the coherence duration of the illumination light; and the first location lies both upstream of the division of the illumination light into the illumination light partial beams and upstream of the retardation component the partial beam components.

13. The illumination optical unit of claim 1, further comprising a scanning device configured to scan an illumination beam over the first facets so that that the illumination light illuminates the illumination field during operation of the illumination optical unit, wherein: when incident on the first facet mirror, an entire beam crosssection of the illumination beam is greater than 20% of a reflection surface of one of the first facets; and the scanning device is configured so that at any given time a given field point in the illumination field is illuminated via at most one of the first facets.

14. The illumination optical unit of claim 13, wherein the scanning device is configured so that a scanning region extending over a plurality of first facets is scanned on the facet mirror during operation of the illumination optical unit.

15. The illumination optical unit of claim 1, further comprising a scanning device configured to scan an illumination beam over the first facets so that the illumination light illuminates the illumination field during operation of the illumination optical unit, wherein: when incident on the first facet mirror, an entire beam crosssection of the illumination beam is greater than 200% of a reflection surface of one of the first facets; and the scanning device is configured so that at any given time a given field point in the illumination field is illuminated only via first facets in which the optical path length difference of their corresponding illumination chancels differs by at least one coherence length.

16. The illumination optical unit of claim 1, wherein the illumination optical unit is configured so that during operation of the illumination optical unit: illumination channel pairs exist in which illumination light guided along the illumination channel pairs is incident at at least one point of the illumination field so that the time difference is less than a coherence duration of the illumination light; and a cross-sectional area of at least one of the illumination channels of the illumination channel pairs does not contribute to illuminating the illumination field.

17. An optical system, comprising: an EUV light source; and an illumination optical unit according to claim 1.

18. An apparatus, comprising: an EUV light source; an illumination optical unit according to claim 1; and a projection optical unit configured to image an object field, which is arranged in the illumination field, into an image field, wherein the apparatus is a projection exposure apparatus.

19. The apparatus of claim 18, further comprising a second facet mirror downstream of the first facet mirror in a beam path of the illumination light through the illumination optical unit, the second facet mirror comprising a plurality of second facets, wherein during operation of the illumination optical unit: each illumination channel is guided by a respective second facet; no more than one illumination channel is guided by a given second facet.

20. A method of using a projection exposure apparatus comprising an illumination optical unit and a projection optical unit, the method comprising: using the illumination optical unit to illuminate a lithography mask; and using the projection optical unit to image at least part of the mask onto a light-sensitive layer, wherein the illumination optical unit is an illumination optical unit according to claim 1.

Description

(1) Exemplary embodiments of the invention will be explained in more detail below on the basis of the drawing. In detail:

(2) FIG. 1 schematically shows a meridional section through a projection exposure apparatus for EUV projection lithography;

(3) FIG. 2 very schematically shows a beam path of a total of six illumination channels for illumination light partial beams for illuminating an illumination field, in which a lithography mask can be arranged, in a further embodiment of a projection exposure apparatus;

(4) FIG. 3 shows a sequence in time of times of impingement of illumination light partial beams on the lithography mask;

(5) FIG. 4 shows a section through an optical retardation component, which is used in an illumination optical unit of the projection exposure apparatus according to FIG. 2;

(6) FIG. 5 shows a sequence in time of times of impingement of illumination light partial beams on the lithography mask when using the optical retardation component according to FIG. 4;

(7) FIG. 6 shows, in an illustration similar to FIG. 3, a sequence in time of times of impingement of illumination light partial beams on the lithography mask;

(8) FIG. 7 shows, in an illustration similar to FIG. 4, a further embodiment of an optical retardation component;

(9) FIG. 8 shows a sequence in time of times of impingement of illumination light partial beams on the lithography mask when using the optical retardation component according to FIG. 7;

(10) FIG. 9 shows, schematically and in relation to an illumination optical unit, a further embodiment of a projection exposure apparatus for EUV projection lithography in a meridional section;

(11) FIG. 10 shows, schematically and in relation to a field facet mirror, some components of an illumination system of the projection exposure apparatus according to FIG. 9 in a plan view, comprising a scanning device for deflecting an EUV output beam of an EUV light source;

(12) FIG. 11 shows, likewise schematically and in relation to a field facet mirror, a further embodiment of a scanning device for deflecting the EUV output beam from the EUV light source, in a cross section;

(13) FIG. 12 shows a plan view of a further embodiment of a field facet mirror and of an illumination field, wherein one field point is highlighted in the illumination field and wherein those associated points, which are imaged on the highlighted illumination field point, on facets of the field facet mirror are also highlighted;

(14) FIG. 13 shows, in an illustration similar to FIG. 12, a plan view of the field facet mirror and of the illumination field according to FIG. 12, wherein an imaging point assignment as in FIG. 12 is specified for exactly one of the field facets and the illumination field and wherein, additionally, respectively one facet region is specified on all other field facets and wherein the mutually superposing images of these facet regions are likewise depicted in the illumination field;

(15) FIGS. 14 and 15 show two examples for sequential scanning of in each case continuous scanning regions on the field facet mirror, wherein the scanning region areas of these scanning regions are slightly smaller in each case than the area of one of the facets;

(16) FIGS. 16 and 17 show variants of scanning region areas over the field facets, which extend over a plurality of field facets;

(17) FIG. 18 shows, schematically and in relation to an illumination optical unit, a further embodiment of a projection exposure apparatus for EUV projection lithography in a meridional section;

(18) FIG. 19 schematically shows the main components of an illumination system of the projection exposure apparatus according to FIG. 18;

(19) FIG. 20 shows, in a schematic plan view approximately corresponding to the viewing direction XX in FIG. 19, three selected pupil facets of the pupil facet mirror of the illumination optical unit and of a reticle of the projection exposure apparatus according to FIGS. 18 and 19 for highlighting optical path length differences of illumination channels within the illumination optical unit;

(20) FIG. 21 shows, in a diagram, a field profile of optical path lengths between the pupil facets according to FIG. 20 over the field coordinate x of the reticle;

(21) FIG. 22 shows, in an illustration similar to FIG. 20, two selected pupil facets and the reticle in one arrangement variant, in which the pupil facet mirror is tilted in such a way that a normal on the main mirror surface of the pupil facet mirror is arranged at a tilt with respect to a plane of incidence of the illumination light partial beams on a main mirror surface of the field facet mirror;

(22) FIGS. 23 to 26 show diagrams, in the style of FIG. 21, in which field profiles of optical path lengths are plotted between various pupil facets and the reticle in the arrangement of the pupil facet mirror in relation to the reticle according to FIG. 22 for various tilt angles of the pupil facet mirror;

(23) FIG. 27 shows, in a diagram, a curvature of a main field surface of the pupil facet mirror, which leads to a parabolic pupil facet mirror-reticle distance function, wherein this distance function is depicted with two different scales;

(24) FIG. 28 shows, in a diagram similar to FIG. 21 and FIGS. 23 to 26, the field profile of optical path lengths between various pupil facets of a pupil facet mirror with a main mirror surface, curved in accordance with FIG. 27, over the reticle;

(25) FIGS. 29 to 31 show different embodiments of a field facet of a field facet mirror which can be used in one of the embodiments of the illumination optical units in a projection exposure apparatus, wherein line-like regions on a reflection surface of the field facet carry an illumination light-blocking coating, wherein FIG. 29 shows a configuration of the blocking coating regions optimized for a y-dipole illumination setting, FIG. 30 shows a corresponding configuration optimized for an x-dipole illumination setting and FIG. 31 in turn shows a corresponding configuration optimized for a plurality of different illumination settings, in which this field facet can be used for reflecting guidance of the illumination light;

(26) FIG. 32 shows, in an illustration similar to FIGS. 14 to 17, a further example for sequential scanning of a continuous scanning region on a further embodiment of a field facet mirror;

(27) FIG. 33 schematically shows a side view of reflection surfaces of the field facets of the field facet mirror according to FIG. 32, as seen from the viewing direction for XXXII in FIG. 32; and

(28) FIG. 34 shows a schematic plan view showing both the field facet mirror according to FIG. 32 and an arrangement region for an associated pupil facet mirror, wherein four illumination channels are highlighted in an exemplary manner.

(29) FIG. 1 schematically shows a meridional section of a projection exposure apparatus 1 for microlithography. The projection exposure apparatus 1 has a light or radiation source 2. An illumination system 3 of the projection exposure apparatus 1 comprises an illumination optical unit 4 for exposing an illumination field, coinciding with an object field 5, in an object plane 6. The illumination field may also be larger than the object field 5, in particular transversely to a displacement direction of the object holder 8. The illumination field may also be smaller than the object field 5, in particular along a displacement direction of the object holder 8. Here, an object in the form of a reticle 7 arranged in the object field 5 is exposed, which reticle is held by an object or reticle holder 8. The reticle 7 is also referred to as a lithography mask. The object holder 8 can be displaced along a displacement direction via an object displacement drive 9. A projection optical unit 10 serves for imaging the object field 5 in an image field 11 in an image plane 12. A structure on the reticle 7 is imaged on a light-sensitive layer of a wafer 13 arranged in the region of the image field 11 in the image plane 12. The wafer 13 is held by a wafer holder 14 (likewise not depicted here). The wafer holder 14 can likewise be displaced along the displacement direction, synchronized with the object holder 8, via a wafer displacement drive 15.

(30) The radiation source 2 is an EUV radiation source with emitted used radiation in the range between 5 nm and 30 nm. Here, this can be a radiation source based on a synchrotron or a free electron laser (FEL). A plasma source, for example a GDPP (gas discharge-produced plasma) source or an LPP (laser-produced plasma) source, may also be used as the radiation source 2. A person skilled in the art finds information in respect of such a radiation source in e.g. U.S. Pat. No. 6,859,515 B2. The light source 2 operates in a pulsed fashion, i.e. emits a sequence in time of light or radiation pulses. A temporal distance T between two light pulses adjacent in time is inversely related to the repetition rate of the light source 2, which e.g. lies in the range from 1 to 100 kHz, e.g. at 50 kHz. The duration of a light pulse is much shorter than the distance T between two light pulses adjacent in time, i.e. the pauses during which no EUV radiation is emitted are much longer than the time periods during which EUV radiation is emitted. The coherence time or the coherence duration emerges from the used spectral bandwidth of the EUV radiation and, in turn, is much shorter than the duration of the light pulse. The used spectral bandwidth can, in particular, be smaller than the spectral bandwidth present at the location of the light source 2.

(31) EUV radiation 16, which emanates from the radiation source 2, is focused by a collector 17. A corresponding collector is known from EP 1 225 481 A. After the collector 17, the EUV radiation 16 propagates through an intermediate focus plane 18 before it impinges on a field facet mirror 19. The field facet mirror 19 is a first facet mirror of the illumination optical unit 4. The field facet mirror 19 has a plurality of field facets, which are not depicted in FIG. 1. The field facets in turn can respectively be subdivided into a plurality of individual mirrors, which is likewise not depicted in the drawing. The field facet mirror 19 is arranged in a plane of the illumination optical unit 4 which is optically conjugate to the object plane 6.

(32) In the following text, the EUV radiation 16 is also referred to as illumination light or imaging light.

(33) After the field facet mirror 19, the EUV radiation 16 is reflected by a pupil facet mirror 20. The pupil facet mirror 20 is a second facet mirror of the illumination optical unit 4. The pupil facet mirror 20 is arranged in a pupil plane of the illumination optical unit 4, which is optically conjugate to the intermediate focus plane 18 and to a pupil plane of the projection optical unit, or which coincides with this pupil plane. The pupil facet mirror 20 comprises a plurality of pupil facets, which are not depicted in FIG. 1. Field facets 25 of the field facet mirror 19 are imaged into the object field 5 with the aid of the pupil facets of the pupil facet mirror 20 and a subsequent imaging optical assembly in the form of a transmission optical unit 21, comprising mirrors denoted by 22, 23 and 24 in the sequence of the beam path. The last mirror 24 of the transmission optical unit 21 is a grazing incidence mirror. Respectively one illumination channel, which guides a partial beam of the illumination light 16, is predetermined by a first facet, i.e. by one of the field facets of the field facet mirror 19, and an associated second facet, i.e. by one of the pupil facets of the pupil facet mirror 20.

(34) In order to simplify the description of positional relationships, FIG. 1 plots a Cartesian xyz-coordinate system as a global coordinate system for the description of the positional relationships of components of the projection exposure apparatus 1 between the object plane 6 and the image plane 12. The x-axis extends perpendicular to the plane of the drawing and into the latter in FIG. 1. In FIG. 1, the y-axis extends to the right and parallel to the displacement direction of the object holder 8 and of the wafer holder 14. In FIG. 1, the z-axis extends downward, i.e. perpendicular to the object plane 6 and to the image plane 12.

(35) The x-dimension over the object field 5 or the image field 11 is also referred to as field height.

(36) FIG. 2 shows a variant of an illumination optical unit 26, which can be used in the projection exposure apparatus 1 in place of the illumination optical unit 4. Components which correspond to those already explained above with reference to FIG. 1 are denoted by the same reference signs and will not be discussed again in detail. Initially, the description below relates to the guidance of the illumination light 16 without the influence of an optical retardation component, which is yet to be explained in more detail below.

(37) FIG. 2 shows a beam path of six illumination channels 27.sub.1, 27.sub.2, 27.sub.3, 27.sub.4, 27.sub.5 and 27.sub.6, by which the partial beams 16.sub.k (k=1, . . . 6) of the illumination light 16 are guided. What is depicted is the beam path between the light source 2 and the reticle 7, wherein the beam path is merely indicated schematically between the light source 2 and the intermediate focus plane 18. The optical path between light source 2 and intermediate focus plane 18 is approximately constant for the illumination light 16. The field facet mirror 15 decomposes the illumination light 16 into partial beams, which are guided along the illumination channels 27.sub.1 to 27.sub.6. That field facet 25, at which the illumination light partial beam of the illumination channel 27.sub.1 is reflected, is denoted as field facet 25.sub.1 below, wherein, for the further illumination channels 27.sub.2 to 27.sub.6, corresponding indexing is undertaken for both the field facets 25 and for the pupil facets 28. By way of example, the field facet 25.sub.3 and the pupil facet 28.sub.3 belong to the illumination channel 27.sub.3. The partial beams 16.sub.k of the illumination light 16 are indexed accordingly below, and so, for example, the illumination light partial beam 16.sub.3 belongs to the illumination channel 27.sub.3.

(38) Depending on the differences in the path lengths of the beam paths, different travel times emerge for the illumination light partial beams 16.sub.k, which are guided over the illumination channels 27.sub.1 to 27.sub.6. Each illumination light partial beam 16.sub.k on its own generates illumination of the illumination field 5, which is not regular but contains a quasi-random intensity variation, referred to as speckle. The speckle contrast increases as the number of transverse modes of the light source decreases. If all modes have the same intensity, the relative speckle contrast equals the inverse square root of the mode number. More information on speckles can be found in Joseph W. Goodman, Speckle Phenomena in Optics, Roberts & Company 2010.

(39) If the path length difference between two illumination channels 27 is smaller than the coherence length of the illumination light 16, wherein the coherence length emerges from multiplying the coherence time of the illumination light by the speed of light, the associated illumination light partial beams 16.sub.k may interfere with one another. In the best case, the residual contrast of the combined illumination of the illumination field 5 by two or more illumination light partial beams 16.sub.k then equals that which would result in the case of illumination by only one partial beam. Particularly in the case of a light source 2 with only a few modes, there may additionally be a systematic interference structure in the illumination on the illumination field 5, which may even lead to a stronger contrast than in the case of a single illumination light partial beam 16. An example of this is formed by the superposition of two partial beams 16 at a small angle, which leads to periodic full modulation.

(40) If the travel time difference between two illumination channels 27 is greater than the coherence length of the illumination light 16, the two independent speckle patterns of the two illumination channels superpose, leading to an improvement in the regularity of the illumination of the illumination field 5. If the travel time difference between two illumination channels 27 is greater than the coherence length of the illumination light 16, the superposition in the illumination field 5 therefore leads to an advantageous behavior, while a travel time difference less than the coherence length leads to no improvement or even to a disadvantageous modification of the regularity.

(41) The differences between the travel times of the illumination light partial beams 16 along the illumination channels 27.sub.1 to 27.sub.6 is therefore selected in such a way in the advantageous embodiment depicted in FIG. 2 that the travel time along all pairs of illumination channels 27.sub.k, 27.sub.l differs in each case by more than the coherence time of the illumination light. What this implies is that the minimum travel time difference between the illumination channel 27 with the shortest path length and the illumination channel with the longest path length, measured between a light source 2 for the illumination light 16 and the illumination field 5, at least equals the product of the illumination channel number and the coherence time of the illumination light. Since the travel time differences along the illumination channels 27.sub.1 to 27.sub.6 depend on the location of incidence in the illumination field 5 and the aforementioned condition must hold for each location of incidence, the travel time difference between the illumination channel with the shortest path length and the illumination channel with the longest path length may be substantially greater than the product of the illumination channel number and the coherence time of the illumination light 16 for at least some field locations, in particular also for almost all field locations.

(42) FIG. 3 clarifies the effect of this configuration of the illumination optical unit 4 or of the illumination optical unit 26. Times of incidence on the reticle 7 of illumination light components 16.sub.1 to 16.sub.6, which were guided along the illumination channels 27.sub.1 to 27.sub.6 and which simultaneously left the light source 2, are depicted along the time axis t. Therefore, absolute time specifications on the depicted time axis t have no meaning; only time differences are of relevance. The width of the plotted bar represents the coherence time or coherence duration .sub.k of the illumination light 16. Overall pulse durations of the illumination channel partial light pulses 16.sub.k can be significantly longer than the bars plotted in FIG. 3. These overall pulse durations of the illumination channel partial light pulses 16.sub.k play no role for the following consideration of interference.

(43) The illumination optical units 4 and 26 are configured in such a way that the coherence duration bars of the partial light pulses 16.sub.k do not overlap.

(44) Therefore, the regularity of the illumination of an illumination field is optimized by such an embodiment of the illumination optical unit since each used illumination channel 27.sub.k leads to an improvement in the regularity.

(45) The configuration of the illumination optical unit in accordance with the described exemplary embodiments can ensure that a travel time difference greater than the coherence length is obtained for all pairs of illumination channels 27.sub.k, 27.sub.l. This results in a corresponding improvement in the regularity of the field illumination. These advantages of the described embodiments become important, particularly in the case of illumination light sources comprising less than 10 transverse modes and having a correspondingly large transverse coherence length. The transverse coherence length must be distinguished from the coherence length emerging by multiplying the coherence time of the illumination light 16 by the speed of light. Furthermore, the advantages of the described embodiments are particularly important in the case of broadband light sources. Finally, the advantages of the described embodiments become important in illumination optical units comprising less than several hundred illumination channels, in particular less than one hundred illumination channels.

(46) What can be avoided by the described embodiments is a too short travel time difference between an illumination channel with a shortest travel time or path length within the illumination optical unit and an illumination channel with a longest path length within the illumination optical unit such that, in particular, disturbing speckle generation could not be precluded in that case. Such a too short travel time difference between the illumination channel with the shortest path length and the illumination channel with the longest path length, which is avoided by the described embodiments, emerges as product of a number of the illumination channels and the coherence length.

(47) FIG. 2 shows the configuration of the illumination optical unit 26 using an optical retardation component 29. The optical retardation component 29 splits the incident illumination light 16 into a plurality of parts with different retardation, which are spatially superposed in the object field 5. The travel time differences between the travel times of the illumination light partial beams along the illumination channels 27.sub.1 to 27.sub.6 are composed in this case from the travel time differences emerging from the different beam paths of the illumination channels 27.sub.1 to 27.sub.6 and the additional retardation travel time differences which the illumination light partial beams experience due to the effect of an optical retardation component 29. The illumination light partial beam 16.sub.k along the illumination channel 27.sub.k therefore consists of a plurality of components 16.sub.k.sup.l, which each have their own travel time, wherein the index k denotes the illumination channel and the index l denotes the retardation by the optical retardation component 29.

(48) In the embodiment according to FIG. 2, the optical retardation component 29 is embodied as a reflecting stepped mirror, which is arranged in the intermediate focus plane 18, i.e. at the location of an intermediate focus between the light source 2 and the field facet mirror 19. In the illumination optical unit 26 according to FIG. 2, this intermediate focus is generated by a collector, not depicted in FIG. 2, in the beam path of the illumination light 16 after the light source 2.

(49) FIG. 4 shows the optical retardation component 29 in more detail. A reflection surface 30 of the optical retardation component 29 comprises a total of four mirror steps 31.sub.1, 31.sub.2, 31.sub.3 and 31.sub.4 in the embodiment according to FIG. 4. Neighboring mirror steps 31 are at a distance of l/2 from one another along a normal N on the reflection surface 30. Since an angle of incidence of the illumination light 16 on the optical retardation component 29 is small, a retardation travel time difference which corresponds to an optical path length of l applies to a good approximation between the components of the illumination light 16 which are reflected at adjacent mirror steps 31.sub.1, 31.sub.2, 31.sub.3 and 31.sub.4. The combination of the possible radiation paths over the four mirror steps 31.sub.1, 31.sub.2, 31.sub.3 and 31.sub.4 with the six illumination channels 27.sub.1, 27.sub.2, 27.sub.3, 27.sub.4, 27.sub.5 and 27.sub.6 results in twenty-four illumination light partial beams 16.sub.1.sup.1 to 16.sub.6.sup.4. Here, the indexing 16.sub.k.sup.l is as already explained above in conjunction with FIG. 2. l is the time analog to the path length difference l, i.e. the travel time difference.

(50) In an illustration analogous to FIG. 3, FIG. 5 clarifies the effect of the optical retardation component 29. Times of incidence of illumination light components 16.sub.k.sup.l on the reticle 7, which have respectively left the radiation source at a given time, are plotted along a time axis t. As a result of the combined effect of, firstly, the different paths of the illumination channels 27 and, secondly, the additional retardation travel time difference due to the retardation component 29, an overall travel time difference emerges for the depicted six illumination channel partial light pulses 16.sub.1 to 16.sub.6 which leads to the time interval, t, between the times of incidence of two illumination channel partial light pulses 16.sub.k.sup.l incident adjacent in time on the reticle 7, which is always greater than a coherence duration of the illumination light 16 from the light source 2. The times of incidence correspond to the left-hand edges of the bars depicted in FIG. 5, while the widths of the bars, analogously to FIG. 3, equal the coherence duration of the illumination light 16. Thus, t>.sub. applies, where .sub. is the coherence duration of the light source 2. In the case of a free electron laser (FEL) as a light source 2 with a corresponding spectral bandwidth, a coherence length, which corresponds to the coherence duration .sub., is e.g. 1.35 m. This applies to a central wavelength of 13.5 nm and a bandwidth / of 100.

(51) Thus, a comparatively large retardation is impressed by the optical retardation component 29. The retardation by the optical retardation component 29 is of the order of magnitude of a retardation step between two adjacent mirror steps 31.sub.1 to 31.sub.4. This step dimension is at least l/2 in the case of the optical retardation component, where l>.sub. applies. Here l is the maximum path length difference between the illumination channel 27 with the shortest path length and the illumination channel with the longest path length. .sub. is the path length analog to the coherence duration .sub..

(52) In the case of the illumination according to FIG. 5, assuming an extreme time resolution and an extremely short light pulse from the light source 2, the reticle sees illumination components 16 from the directions of the illumination light channels 27.sub.1, 27.sub.2, 27.sub.5, 27.sub.4, 27.sub.6 and 27.sub.3 in quick succession. This light showering from different directions repeats a number of times in quick succession, to be precise once per used step of the optical retardation component 29.

(53) Analogously to FIG. 3, FIG. 6 shows, along a time axis t, times of incidence of illumination light components 16.sub.1 to 16.sub.6, which were guided along the illumination channels 27.sub.1 to 27.sub.6 and left the light source 2 at a given time, on the reticle 7. The main difference between FIG. 3 and FIG. 6 is that FIG. 6 is based on an illumination optical unit 4 or 26, in which the path length differences between the illumination light channels 27.sub.1 to 27.sub.6 are greater than in the embodiment according to FIG. 3.

(54) FIG. 7 shows a further embodiment of an optical retardation component 33. It is described below where it differs from the optical retardation component 29. A main difference from the optical retardation component 29 lies in the order of magnitude of a retardation step between two adjacent mirror steps 31.sub.1 to 31.sub.4. In the optical retardation component 33, this step dimension is l/2, where .sub.<l<L/M applies. Here, L is the minimum path length difference along two illumination light channels 27.sub.1 to 27.sub.6 and M is the number of steps in the retardation component 33.

(55) Thus, a comparatively small retardation is impressed by the optical retardation component 33, and so all components 16.sub.i.sup.l of the illumination light 16, which left the radiation source 2 at a given time and ran over different steps of the retardation component 33 via a specific illumination light channel 27.sub.i, reached the reticle 7 before components 16.sub.j.sup.l of the illumination light 16, which passed over a different illumination channel 27.sub.j, reach the reticle 7. This is depicted in FIG. 8 in a representation analogous to FIG. 5.

(56) In the sequence of incidence in time according to FIG. 8, assuming an extreme time resolution and an extremely short light pulse from the light source 2, the reticle 7 initially sees a sequence of components, corresponding to one another in time, of the illumination channel partial light pulses from a first illumination direction, then a sequence of components, corresponding to one another in time, of the illumination channel partial light pulses from a second illumination direction, and so on.

(57) A minimum time interval t between the times of incidence of the components, corresponding to one another in time, of the illumination channel partial light pulses 16.sub.k.sup.l on the reticle 7 is also greater than the coherence duration .sub. of the light source 2 in this case. l is the time analog to the path length difference l, i.e. the travel time difference. Here, l=t applies.

(58) The optical retardation component 33 can be arranged at the location of the pupil facets of the pupil facet mirror 22. This avoids an additional reflection and the light loss connected therewith. As a result of the small step height, an easier integration into other optical components such as e.g. a facet mirror is possible than is the case where there are larger step heights.

(59) The optical retardation component 29 can be arranged at the location of an intermediate focus. Large step heights are slightly more complicated to produce, and only a single such element is required in the case of placement at the location of an intermediate focus.

(60) In the case of the projection exposure with the aid of the projection exposure apparatus 1, at least part of the reticle 7 in the object field 5 is imaged on a region of the light-sensitive layer on the wafer 13 in the image field 11 for the lithographic production of a microstructured or nanostructured component, in particular a semiconductor component, for example a microchip. Here, the reticle 7 and the wafer 13 are displaced synchronized in time and continuously in the y-direction during the scanning operation.

(61) In the following text, a further embodiment of a microlithographic projection exposure apparatus will be explained. Components and functions which were already explained above are denoted by the same reference signs and are only explained below where this is necessary for the understanding of deviating components or functions.

(62) The microlithographic projection exposure apparatus 1 according to FIG. 9 serves for producing a microstructured or nanostructured electronic semiconductor component. The light or radiation source 2 emits EUV radiation in the wavelength range of, e.g., between 5 nm and 30 nm. In the embodiment according to FIG. 9, the light source 2 is embodied as a free electron laser (FEL). Here, this is a synchrotron radiation source which generates coherent radiation with very high brilliance. Prior publications, in which such FELs are described, are specified in WO 2009/121 438 A1. A light source 2, which can be used, for example, in the embodiment according to FIG. 9, is described in Uwe Schindler Ein supraleitender Undulator mit elektrisch umschaltbarer Helizitt [A Superconducting Undulator having Electrically Switchable Helicity], Research Center Karlsruhe in the Helmholtz Association, Scientific Reports, FZKA 6997, August 2004, and in DE 103 58 225 B3.

(63) The EUV light source 2 has an electron beam supply device 2a for generating an electron beam 2b and an EUV generation device 2c. The latter is supplied with the electron beam 2b via the electron beam supply device 2a. The EUV generation device 2c is embodied as an undulator.

(64) The light source 2 has a mean power of 2.5 kW. The pulse frequency of the light source 2 is 30 MHz. Each individual radiation pulse then carries an energy of 83 J. In the case of a radiation pulse length of 100 fs, this corresponds to a radiation pulse power of 833 MW.

(65) A used radiation beam, which is also referred to as an output beam, is used as illuminating or imaging light 16 for the purposes of illuminating and imaging within the projection exposure apparatus 1. The used radiation beam 16 is illuminated with the aid of a scanning device 36, yet to be described below, within an aperture angle 35 which is matched to the illumination optical unit 4 of the projection exposure apparatus 1. Emanating from the light source 2, the used radiation beam 16 has a divergence which is less than 5 mrad. The scanning device 36 is arranged in the intermediate focus plane 18 of the illumination optical unit 4. After the scanning device 36, the used radiation beam 16 is initially incident on the field facet mirror 19. Details in respect of the scanning device 36 are yet to be explained in more detail below on the basis of FIGS. 10 and 11.

(66) In particular, the used radiation beam 16 has a divergence which is less than 2 mrad and preferably less than 1 mrad. The spot size of the used radiation beam on the field facet mirror 19 is approximately 4 mm.

(67) In an exemplary manner, FIG. 10 shows a facet arrangement, a field facet array, of field facets 25 of the field facet mirror 19. The field facet mirror 19 represents a facet mirror of an illumination system of the projection exposure apparatus 1. Only some of the actually present field facets 25 are depicted in an exemplary manner in three columns and 15 rows. The field facet array of the field facet mirror 19 has a total of 6 columns and 75 rows. The field facets 25 have a rectangular shape. Other shapes of the field facets 25 are also possible, for example an arcuate shape, as is depicted below in conjunction with FIGS. 12 to 17, or a ring-shaped or partial ring-shaped geometry. Overall, the field facet mirror 19 therefore has 450 field facets 25 in one possible variant. Each field facet 25 has an extent of 50 mm in the horizontal direction in FIG. 10 and of 4 mm in the vertical direction in FIG. 10. Accordingly, the overall field facet array has an extent of 300 mm300 mm. The field facets 25 are not depicted true to scale in FIG. 10.

(68) After reflection on the field facet mirror 19, the used radiation beam 16, which is split into the pencil of rays or the illumination light partial beam assigned to the individual field facets 25, is incident on the pupil facet mirror 20. Pupil facets 28 of the pupil facet mirror 20, of which FIG. 9 merely depicts one pupil facet 28, are round. Each pencil of rays of the used radiation beam 16 reflected from one of the field facets 25 is assigned to one of these pupil facets 28, and so respectively one impinged facet pair with one of the field facets 25 and one of the pupil facets 28 predetermines an illumination channel or beam guidance channel 27 for the associated pencil of rays of the used radiation beam 16. The channel-by-channel assignment of the pupil facets 28 to the field facets 25 is brought about dependent on a desired illumination by the projection exposure apparatus 1. The output beam 16 is therefore guided along the illumination channel, in sequence over pairs of respectively one of the field facets 25 and respectively one of the pupil facets 28, for predetermining individual illumination angles. It is also possible to simultaneously illuminate a plurality of the field facets 25 and hence a plurality of illumination channels 27. At any given time, the illumination light 16 can be guided over exactly one pair, or else over a plurality of, but generally few, pairs, made of respectively one field facet 25 and respectively one pupil facet 28. The field facets 25 are in each case tilted individually for actuating respectively predetermined pupil facets 28.

(69) The field facets 25 are imaged into the illumination or object field 5 in the reticle or object plane 6 of the projection optical unit 10 of the projection exposure apparatus 1 by the pupil facet mirror 20 and the subsequent transmission optical unit 21 consisting of the three EUV mirrors 22, 23, 24. The EUV mirror 24 is embodied as a grazing incidence mirror.

(70) An illumination angle distribution of the illumination of the object field 5 by the illumination optical unit 4 emerges from the sequence of the individual illumination angles predetermined by the illumination channels of individual facets pairs 25, 28 via the scanning integration of all illumination channels, brought about by a scanning illumination of the facets 25 of the field facet mirror 19 with the aid of the scanning device 36.

(71) In the case of an embodiment of the illumination optical unit 4 (not depicted here), in particular in the case of a suitable position of an entry pupil of the projection optical unit 10, it is also possible to dispense with the mirrors 22, 23 and 24, leading to a corresponding transmission increase of the projection exposure apparatus for the used radiation beam 16.

(72) An overall dose of 49.2 J arrives at the whole object field 5 in each complete scan of the field facet mirror 19. In order to obtain the overall dose in the image field 11, this overall dose still needs to be multiplied by the overall transmission of, firstly, the illumination optical unit 4 and, secondly, the projection optical unit 10. These exemplary specifications assume that a complete scan of the whole field facet mirror 19 is performed during the period of time which a point on the wafer 13 requires to pass through the image field 11. In principle, faster scanning of the illumination light 16 over the field facet mirror 19 is possible. The period of time which a point on the wafer 13 requires to pass through the image field 11 in the case of scanning exposure may be an integer multiple of the period of time required to scan the whole field facet mirror 19 with the illumination light 16.

(73) The reticle 7 reflecting the used radiation beam 16 is arranged in the object plane 6 in the region of the object field 5. The reticle 7 is supported by the reticle holder 8 which can be displaced in an actuated manner via the reticle displacement drive 9.

(74) The projection optical unit 10 images the object field 5 in the image field 11 in the image plane 12. During the projection exposure, the wafer 13 carrying a light-sensitive layer is arranged in this image plane 12, which light-sensitive layer is exposed by the projection exposure apparatus 1 during the projection exposure. The wafer 13 is supported by a wafer holder 14, which in turn can be displaced in a controlled manner via a wafer displacement drive 15.

(75) In order to simplify the representation of positional relationships, an xyz-coordinate system is used below. The x-axis extends perpendicular to the plane of the drawing of FIG. 9 and points into the latter. In FIG. 9, the y-axis extends to the right. In FIG. 9, the z-axis extends downward.

(76) During the projection exposure, both the reticle 7 and the wafer 13 are scanned in a synchronized manner in the y-direction in FIG. 9 by an appropriate actuation of the reticle displacement drive 9 and of the wafer displacement drive 15. The wafer is scanned in the y-direction with a scanning speed of typically 600 mm/s during the projection exposure. The synchronized scanning of the two displacement drives 9, 15 may be performed independently of the scanning operation of the scanning device 36.

(77) The long side of the field facets 25 is perpendicular to a scanning direction y. The x/y-aspect ratio of the field facets 25 corresponds to that of the slit-shaped object field 5, which can likewise be embodied in a rectangular or arcuate form.

(78) FIGS. 10 and 11 show the embodiments of the scanning device 36 for the used radiation beam 16 in more detail. In order to simplify the representation of positional relationships, an xy-coordinate system is used for the scanning device in FIG. 10. The x-axis, which is parallel to the x-axis, extends into the plane of the drawing in FIG. 10. The y-axis, which lies in the yz-plane, extends obliquely toward the top right in FIG. 10.

(79) In order to represent positional relationships in relation to the field facet mirror 19, an x.sub.FFy.sub.FF-coordinate system is used accordingly. The x.sub.FF-axis extends parallel to the x-axis, i.e. in the direction of the longer sides of the rectangular field facets 25. The y.sub.FF-direction extends perpendicular thereto in the direction of the shorter sides of the rectangular field facets 25. In the x.sub.FF-direction, the field facets 25 have an extent of X.sub.FF. In the y.sub.FF-direction, the field facets 25 have an extent of Y.sub.FF. X.sub.FF/Y.sub.FF is the x/y-aspect ratio of the field facets 25.

(80) In the embodiment according to FIG. 10, the scanning device 36 is a scanning mirror reflecting the used radiation beam 16 in a grazing manner, which scanning mirror can be tilted about a line scan axis 37, which coincides with the y-axis, and about a line feed axis 38, which is parallel to the x-axis. Both axes 37, 38 lie in a reflecting mirror surface 39 of the scanning device 36.

(81) In FIG. 11, the field facet mirror 19 is depicted schematically as a 44 array with four horizontal rows, each having four field facets 25. The following frequency and time data relate to the illumination of the field facet mirror 19, already described in conjunction with FIG. 10, with the 675 array. In the described embodiment, the tilting about the line scan axis 37 for scanning a field facet line along the x.sub.FF-direction occurs with a line frequency of 7.5 kHz. Here, the mirror surface 39 is tilted by +/4.5, leading to a deflection angle for the used radiation beam 16 of +/9. Accordingly, the dwell time of the used radiation beam 16 on respectively one line (y.sub.FF=const) of the field facet mirror 19 is 133.3 s. The line feed in the y.sub.FF-direction is brought about by synchronized tilting about the line feed axis 38 such that the 75 lines are scanned with the correct line distance, wherein the tilting about the line feed axis 38 also ensures a return of the used radiation beam 16 to the first field facet 25a to be scanned from the last scanned field facet 25z. Therefore, the mirror surface 39 is additionally tilted about the line feed axis 38 with a frequency of 100 Hz. The dwell time per individual field facet 25 is 22.2 s. Thus, 660 EUV radiation pulses are incident on the field facet 25 during the dwell time on one of the field facets 25.

(82) When scanning the field facet mirror 19, the illumination on the field facet mirror 19 can be moved continuously in the y-direction. Such a scanning movement can be achieved with mechanically comparatively simple and durable components.

(83) A higher line frequency than the above-described line frequency of 7.5 kHz is also possible, for example a line frequency of 10 kHz, of 15 kHz, of 20 kHz or else an even higher line frequency.

(84) The distance between the mirror surface 39 and the field facet mirror 19 is approximately 1 m.

(85) Instead of tilting about the line scan axis 37, the line scan can also be generated with the aid of a polygon scanner 40, which rotates around the line scan axis 37. This is depicted schematically in FIG. 11, which shows a plan view of the field facet mirror 19. The x.sub.FF-axis extends to the right in FIG. 11 and the y.sub.FF-axis extends perpendicular to the plane of the drawing, toward the observer, in FIG. 11.

(86) For the line scan, the polygon scanner has a polygon mirror 41 with a total of six polygon facets, i.e. it is embodied as a regular hexagon in the circumferential direction around the rotational axis 37 thereof. When the scanning device 36 is embodied with the polygon mirror 41, a tilting mirror is arranged upstream or downstream thereof, which tilting mirror, as described above, can be tilted about the line feed axis 38. Additionally or alternatively, the illumination radiation from the light source 2 can be widened in the y-direction by an optical unit not depicted in FIG. 11. The specified beam diameters then only relate to the x-extent.

(87) When incident on one of the total of six mirror surfaces 39 of the polygon mirror 41, the used radiation beam 16 has a diameter of approximately 5 mm.

(88) A distance between the light source 2 and the polygon mirror 41 is approximately 1 m.

(89) When incident on the field facet mirror 19, i.e. after the reflection on the polygon mirror 41, the used radiation beam 16 has a diameter of approximately 10 mm.

(90) The image field 11 has a slit width parallel to the scanning direction y of 2 mm and a slit width perpendicular to the scanning direction, i.e. in the x-direction, of 26 mm. In the case of an assumed dose of 30 mJ/cm.sup.2 on the wafer 13, which ensures that the light-sensitive layer reacts, a scanning speed of 600 mm/s of the reticle 7 and an image field width of 26 mm, the output beam 16 must arrive on the wafer 13 with a power of 5 W.

(91) The respective embodiment of the scanning device 36 is such that in each case continuous scanning regions are scanned on the field facet mirror 19 in sequence, which scanning regions have a scanning region area corresponding at most to the area of the field facets 25. This will be explained in more detail below on the basis of FIG. 12ff. Components and functions which correspond to those which are respectively described with reference to figures already explained above are denoted by the same reference signs and are not discussed in detail again.

(92) On the left-hand side, FIG. 12 schematically shows a top view of a variant of the field facet mirror 19 with arcuate field facets 25. Three columns of field facets 25 and a total of twenty field facets 25 are depicted. Naturally, the number of field facets 25 in the field facet mirror 19 according to FIG. 12 is much greater in reality, as already explained above.

(93) On the right-hand side, FIG. 12 depicts the illumination field 5, on which the field facets 25 of the field facet mirror 19 are imaged in a manner superposing one another.

(94) In the illumination field 5 according to FIG. 12, one field point 42 is highlighted. In FIG. 12, those facet points 43 which are imaged exactly on the field point 42 (original images of the field point 42) via the transmission optical unit 21 are also highlighted. As a result of the spatial relationships of the image of the facet points 43 on the field point 42, the facet points 43 within the respective contour of the field facets 25 do not all lie at exactly the same relative location, but may be slightly displaced with respect to one another, both in the x.sub.FF-direction and in the y.sub.FF-direction. This displacement, for which a displacement x.sub.FF for the two upper-most field facets 25 of the left-hand column of the field facet mirror 19 according to FIG. 12 is depicted in an exemplary manner, is small compared to the dimensions of the field facets 25 in the x.sub.FF-direction and in the y.sub.FF-direction. Analogous to the displacement x.sub.FF, there can also be a displacement y.sub.FF along the y.sub.FF-direction. This is depicted in FIG. 12 on the basis of the facet points 43 of the two upper-most field facets 25 in the right-hand column. In the y.sub.FF-direction, these facet points 43 have a distance of Y.sub.FF plus y.sub.FF from one another, i.e. are spaced further apart from one another than the extent Y.sub.FF of the field facets 25 in the y.sub.FF-direction by an amount y.sub.FF. Depending on the selection of the field point 42 and the respectively associated facet point 43 of a specific field facet 25, a specific value for y.sub.FF emerges for each one of the field facets 25. When considering all the possible combinations of field points 42 and facet points 43 of the field facets 25, a maximum value y.sub.FF,max, which y.sub.FF can assume, emerges. This dimension y.sub.FF,max is also referred to as maximum original image displacement of a field point 42 on the facet mirror 25.

(95) In an illustration similar to FIG. 12, FIG. 13 in turn shows the field facet mirror 19 and the illumination field 5. For the field point 42, the facet point 43, which is imaged on the field point 42 via the transmission optical unit 21, is specified for one of the field facets 25, namely the field facet 25 third from bottom in the central column of the field facet mirror 19.

(96) For the other field facets 25, FIG. 13 in each case specifies a facet region 44, which, with reference to the x.sub.FF-coordinate and y.sub.FF-coordinate of each individual field facet 25, lies at exactly the same location on the reflection surface of each field facet 25.

(97) The facet region images 45 of these facet regions 44 are not identical, but overlap one another on the illumination field 5 due to the aforementioned geometric deviation when imaging using the transmission optical unit 21 and also due to the slightly different imaging scales of this image resulting therefrom.

(98) The facet region images 45 of the facet regions 44 are depicted on the illumination field 5 on the right-hand side in FIG. 13, which facet region images are generated on the illumination field 5 by the transmission optical unit 21. The field point 42 lies in all facet region images 45. To the extent that the scanning device 36 is actuated in such a way that the facet regions 44 are not illuminated at a given time, the field point 42 can be illuminated by the facet point 43 at the time without this resulting in disturbing interference as a result of illumination of this field point 42 by the other facets 25.

(99) The field point 42 in the right-hand side of FIG. 13 is then only illuminated by one of the facets 25, namely by the facet 25 third from bottom in the central column of the field facet mirror 19 in the left-hand side of FIG. 13.

(100) FIG. 14 likewise shows a top view of the field facet mirror 19. FIG. 14 depicts a variant of an actuation of the scanning device 36, in which a continuous scanning region 46 on the field facet mirror 19 is scanned in sequence, the scanning region area of which continuous scanning region being smaller than the reflection surface of one of the field facets 25. In the actuation variant according to FIG. 14, the scanning region 46 is arcuate like the field facets 25. In the x.sub.FF-direction, the scanning region 46 has the same extent as the field facets 25. In the .sub.yFF-direction, the extent of the scanning region 46 is slightly smaller than the extent of the field facets 25. The y.sub.FF-extent of the field facets 25 is indicated by two dashed delimitation lines 47 in FIG. 14 for comparison purposes. A bounding shape of the scanning region 46 corresponds to the bounding shape of the field facets 25.

(101) A scanning region area S of the scanning region 46 is smaller than the area F=X.sub.FFY.sub.FF of one of the field facets 25 by at least the ratio of, firstly, the maximum original image displacement y.sub.FF,max of a field point 42, i.e. of an object point of the illumination field 5, on the field facet mirror 19 to, secondly, the facet extent Y.sub.FF parallel hereto. Hence, the following applies:
S=F(1(y.sub.FF,max/Y.sub.FF))

(102) In the variant according to FIG. 14, the scanning device 36 is actuated in such a way that the continuous scanning region 46 is always scanned on the field facet mirror 19, wherein this continuous scanning region 46 then is displaced column-by column in the y.sub.FF-direction over the field facet mirror 19 in accordance with a time profile of the actuation of the scanning device 36.

(103) Thus, two scanning steps are superposed when actuating the scanning device 36. Firstly, an area corresponding to the scanning region 46 is always scanned and this scanning region is secondly scanned column-by-column over the field facets 25 of the field facet mirror 19. At the end of one of the columns of the field facet mirror 19, scanning is continued in the adjacent column.

(104) The selected size of the scanning region 46 leads to no field point on the illumination field 5 being illuminated simultaneously by two different field facets 25. Therefore, disturbing interferences of the illumination light on the illumination field points are precluded.

(105) While the scanning region 46 moves over the facet mirror 19, two adjacent field facets 25 are occasionally impinged upon simultaneously by the illumination light 16, as shown in the snapshot according to FIG. 14. However, since only portions of these two adjacent field facets 25 which cause different regions of the illumination field 5 to be impinged by the illumination light 16 are impinged upon, there nevertheless is no illumination light interference in the illumination field 5.

(106) FIG. 15 shows a further variant of an actuation of the scanning device 36 for sequential illumination of the whole field facet mirror 19. Components and functions corresponding to those already explained above with reference to the other figures are denoted by the same reference signs and are not discussed in detail again.

(107) In the embodiment according to FIG. 15, the scanning region 46 is rectangular. An area of the scanning region 46 is smaller than the area of one of the field facets 25. The y.sub.FF-extent of the scanning region 46 according to FIG. 15 is once again slightly smaller than the y.sub.FF-extent of the field facets 25, as already explained in conjunction with FIG. 14. Therefore, the bounding shape of the scanning region 46 does not correspond to the bounding shape of the field facets 25 in the embodiment according to FIG. 15.

(108) The scanning region 46 can not necessarily be written in one of the field facets 25.

(109) Two further examples of the scanning region configuration are explained on the basis of FIGS. 16 and 17, via which respectively one field point in the illumination field 5 is only illuminated by in each case at most one of the facets 25 of the field facet mirror 19. Components and functions corresponding to those which were already explained above with reference to the other figures are denoted by the same reference signs and will not once again be discussed in detail.

(110) In the embodiments according to FIGS. 16 and 17, the scanning device 36 for illuminating the field facet mirror 19 is embodied in such a way that a scanning region 48 extends over a plurality of the facets 25.

(111) In the variant according to FIG. 16, the scanning region 48 is embodied continuously in the form of a scanning range strip extending over three facet columns. This scanning region strip extends obliquely over all three facet columns in the embodiment according to FIG. 16. The gradient of this oblique profile must be so large that a y.sub.FF-offset between adjacent, illuminated facet columns is greater than a y.sub.FF-extent of the scanning region 48. The offset is indicated by a dashed line 49 in FIG. 16 and the y.sub.FF-extent of the scanning region 48 is indicated by a further dashed line 50 in FIG. 16. A difference y in the y.sub.FF-direction between the offset 49 and the y.sub.FF-extent 50 of the scanning region 48 is greater than the maximum original image displacement y.sub.FF,max, which was already explained above in conjunction with FIG. 12. At the same time, the profile of the scanning region 48 according to FIG. 16 must be such that an overall y.sub.FF-extent A of the scanning region 48 is at most as large as the y.sub.FF-extent Y.sub.FF of the individual field facets 25. Here, the difference between the y.sub.FF-extents A of and Y.sub.FF is, in turn, at least as large as the maximum original image displacement y.sub.FF,max. By observing these boundary conditions, what is ensured is that respectively one field point in the illumination field 5 is only illuminated by in each case at most one of the field facets 25, even in the case of an actuation leading to a scanning region 48 according to FIG. 16.

(112) When illuminating the field facet mirror 19, the scanning region 48 is scanned in the y.sub.FF-direction over the field facet mirror 19 in such a way that each one of the field facets 25 is illuminated within one scanning period.

(113) In the actuation variant according to FIG. 17, a plurality of scanning regions 48a, 48b, 48c, the x.sub.FF-extent of each respectively corresponding to one field facet column width, are scanned in the y-direction over the various columns of the field facet mirror 19. In the y.sub.FF-direction, the scanning regions 48a, 48b and 48c respectively have a y.sub.FF-offset from one another which is greater than 0. A y-offset between the scanning regions 48a, 48b and 48c may be greater than a y-extent of the respective scanning region 48a to 48c, plus the maximum original image displacement y.sub.FF,max. Once again, an overall extent A of the scanning region 48, formed from the portions 48a to 48c, according to FIG. 17 is at most the y.sub.FF-extent of the individual field facets 25. Here, once again, the difference between the overall extent A and the y.sub.FF-extent of the individual field facets 25 can lie in the order of magnitude of the maximum original image displacement y.sub.FF,max.

(114) An offset of the scanning region 48 between two column-by-column adjacent field facets 25 is at least the sum of the y.sub.FF-extent of the scanning region 48 and the maximum original image displacement y.sub.FF,max which was already explained above in conjunction with FIG. 12.

(115) If the area of the scanning region 46 or 48 is small, the area of the simultaneously illuminated region of the illumination field 5 is also small. The overall duration, during which a point of the wafer 13 is impinged upon by used light 16, also depends on the ratio of the overall size of the illumination field 5 to the size of the simultaneously illuminated region of the illumination field, in addition to the size of the image field 11 and the scanning speed of the wafer controlled by the wafer displacement drive 15. As the length of this overall duration increases, the smaller the effect, which spatially and temporally localized interferences may have, becomes.

(116) The area of the scanning region 46 or 48 can be as large as possible, without, however, breaking the previously described conditions on the embodiment of the scanning region in the process. The scanning region 46 or 48 can be selected in such a way that the area thereof is greater than 20% of a reflection surface of one of the field facets. The area of the scanning region 46 or 48 can be greater than a third of the area of one of the field facets 25, can be greater than half of the area of one of the field facets 25, greater than 75% of the area of one of the field facets 25 or can be greater than 90% of the area of one of the field facets 25.

(117) In an embodiment of a projection exposure apparatus, not depicted in a drawing, all travel time differences along pairs of illumination channels 27.sub.i, which belong to field facets 25 within the same row of the field facet mirror 19, are longer than the coherence time of the illumination light 16. However, the travel time differences along pairs of illumination channels 27.sub.i, which belong to field facets 25 within different rows of the field facet mirror 19, may be shorter than the coherence time of the illumination light 16. The scanning region 46 has a y-extent as described in conjunction with FIGS. 14 and 15. The x-extent comprises the whole field facet mirror, and so the area of the scanning region 46 now is greater than the area of one field facet 25.

(118) In the following text, a further embodiment of a microlithographic projection exposure apparatus is explained. Components and functions which were already explained above are denoted by the same reference signs and are only mentioned in the following text where this is required for the understanding of deviating components or functions.

(119) In the further embodiment according to FIG. 18, the microlithographic projection exposure apparatus 1 likewise serves for producing a microstructured or nanostructured electronic semiconductor component.

(120) The used radiation beam 16 is illuminated within the aperture angle 35, which is adapted to the illumination optical unit 4 of the projection exposure apparatus 1, with the aid of a beam shaping device 51 yet to be described below. The beam shaping device 51 is embodied as an elliptic mirror. Emanating from the light source 2, the used radiation beam 16 has a divergence which is less than 1 mrad. The beam shaping device 51 is arranged in the intermediate focus plane 18 of the illumination optical unit 4. Downstream of the beam shaping device 51, the used radiation beam 16 is initially incident on the field facet mirror 19.

(121) In the embodiment according to FIG. 18, the used radiation beam 16 more particularly has a divergence which is less than 500 rad and preferably less than 150 rad. The diameter of the used radiation beam 16 on the beam shaping device 51 is approximately 5 mm. When emerging from the light source 2, the used radiation beam 16 has an almost vanishing diameter, which is less than 1 mm and may also be less than 500 m or may be less than 250 m.

(122) The field facet mirror 19 has a field facet array (not depicted in any more detail here). All that is depicted in FIG. 18 in an exemplary manner is merely some of the actually present field facets 25. The field facets 25 are arranged on a main mirror surface of the field facet mirror 19.

(123) The pupil facets 28 are arranged on a main mirror surface of the pupil facet mirror 20.

(124) The pupil facet mirror 20 is arranged in the region of a position of an entry pupil of the projection optical unit 10. The illumination channels for the individual illumination light partial beams lead to the illumination field 5 from the field facets 25.

(125) An illumination angle distribution of the illumination of the object field 5, which is caused by the illumination optical unit 4, emerges from the individual illumination angles, which are predetermined over each facet pair 25, 28, by the integration of all illumination channels caused by an illumination of the facets 25 of the field facet mirror 19 with the aid of the beam shaping device 51.

(126) The entry pupil of the projection optical unit 10 is situated, possibly after imaging by some or all of the mirrors 22, 23 and 24, in the vicinity of the pupil facet mirror 20.

(127) To the extent that individual components of the projection exposure apparatus 1 are described, use is made of a local xy-coordinate system which spans the main reflection surface or main mirror surface of this component. The x-directions of the global coordinate system and of the local coordinate systems generally coincide or are parallel to one another.

(128) The beam shaping device 51 is an elliptically shaped mirror, on which the used radiation beam 16 is incident in a grazing manner. The distance between the light source 2 and the beam shaping device 51 is e.g. 25 m, 50 m or else 100 m. In particular, the distance can depend on the divergence of the light source 2. As a result of this, a spot with a dimension of e.g. 5 mm results on the beam shaping unit 51.

(129) The elliptically shaped mirror of the beam shaping unit 51 leads to imaging of a beam guidance object plane, which is not depicted in any more detail in the drawing, in a beam guidance image plane. The beam guidance image plane substantially corresponds to the field facet mirror 19. As a result of the small divergence of the used radiation beam 16 from the light source 2, the exact position of the beam guidance object plane is irrelevant and, in the beam guidance object plane, the used light 16 forms a spot with a dimension which only deviates insubstantially from the spot dimension on the beam shaping unit 51. The elliptically shaped mirror from the beam shaping device 51 leads to optical imaging with the imaging scale ||=60, and so the used radiation beam 16 illuminates the whole field facet mirror 19.

(130) The distance between the beam shaping unit 51 and the field facet mirror 19 is approximately 2 m.

(131) FIG. 19 schematically shows the main components of the projection exposure apparatus 1. Initially, the light source 2 is imaged in the intermediate focus plane 18 by the collector 17, which is not depicted in FIG. 18. In the case of a light source 2 with a small etendue, in particular in the case of an electron beam-based light source 2 such as e.g. an FEL laser, it is also possible to dispense with the collector 17. The pupil facet mirror 20 with the pupil facets 28 is arranged in such a way that images of the light source 2 come to rest at the location of the pupil facets 28. Therefore, an image of the light source 2 lies in the region of each illuminated pupil facet 28. The pupil facet mirror 20, possibly together with further components of the transmission optical unit 21 (cf. FIG. 18), images the field facets 25 of the field facet mirror 19 in the illumination field or object field 5 in a manner superposed on one another. The optical path length of all rays of the illumination light 16 incident on a specific pupil facet 28 is constant at the location of this pupil facet 28, i.e. the optical path length does not depend on which ray of the illumination light 16, which has reached this pupil facet 28, is considered.

(132) In the following text, a field profile of an optical path length opd (optical path distance) between pupil facets 28 of the pupil facet mirror 20 and the reticle 7 is discussed on the basis of FIGS. 20 and 21. To this end, three pupil facets 28.sub.1, 28.sub.2 and 28.sub.3 of the pupil facet mirror 20, which all three lie in the same xz-plane, are singled out in an exemplary manner. Adjacent ones of these three pupil facets 28.sub.1 to 28.sub.3 have a distance of 125 mm from one another. In the z-direction, the pupil facet mirror 20 is a distance of 1500 mm from the reticle 7. A portion of the reticle 7, considered in FIG. 20, has an extent of 104 mm in the x-direction. The central one of the three considered pupil facets, i.e. the pupil facet 28.sub.2, lies level with the center of the reticle 7 (x=0) in the x-direction.

(133) The diagram according to FIG. 21 highlights dependence of an optical path length between the pupil facets 28.sub.1 to 28.sub.3 on the x-coordinate of the reticle 7. The pupil facet 28.sub.3, depicted all the way on the right in FIG. 20, has the greatest distance from the left-hand edge of the reticle 7 (x=52 mm), which distance continuously reduces toward the right-hand edge of the reticle 7 (x=+52 mm). This is depicted by a path length curve 52 in FIG. 21. A correspondingly mirrored path length curve 53 emerges for the pupil facet 28.sub.1 depicted on the left-hand side in FIG. 20. For the third pupil facet 28.sub.2, arranged centrally over the reticle 7, the path length curve 54 emerges, which has the smallest distance to the center of the reticle 7 (x=0), the distance continuously increasing to the two edges of the reticle 7 (x=+/52 mm).

(134) The two outer pupil facets 28.sub.1 and 28.sub.3 have exactly the same distance from the center of the reticle 7 (x=0). There, the two path length curves 52 and 53 intersect. To the extent that, proceeding from the light source 2, the illumination channels for the respective illumination light partial beams, which have led to the pupil facets 28.sub.1 and 28.sub.3, have the same length to the pupil facets 28.sub.1 and 28.sub.3 from the light source 2, disturbing interference follows from the following path length equality for x=0, which is avoided by an appropriate arrangement of the pupil facet mirror 20, as will be explained below.

(135) In an illustration similar to FIG. 20, FIG. 22 shows the pupil facet mirror 20, which is now arranged with such a tilt that the normal N thereof with respect to a main mirror surface 55 of the pupil facet mirror 20 is tilted by a tilt angle =10 with respect to the yz-plane, i.e. with respect to a plane of incidence of the illumination light partial beam on a main mirror surface 56 of the field facet mirror 19. This tilt leads to those partial beams extending in the yz-plane up to the pupil facet mirror 20 subsequently being deflected out of the yz-plane in accordance with this tilt angle.

(136) FIG. 26 shows the field profile of the path lengths in relation to the reticle 7 in an illustration corresponding to FIG. 21 for this tilt angle of 10 of the pupil facet mirror 20, for both the pupil facets 28.sub.1 and 28.sub.3 and for fourteen further pupil facets 28 arranged equidistantly between these two pupil facets 28.sub.1, 28.sub.3. The pupil facet 28.sub.1 always has the greatest distance from the reticle 7, which is depicted by the uppermost path length curve 57 in FIG. 26. The pupil facet 28.sub.3 always has the smallest distance from the reticle 7 over the whole x-dimension thereof, which is depicted by the lowermost path length curve 58 in FIG. 26. The fourteen pupil facets 28 lying therebetween have field profiles which are depicted by path length curves 59 lying between the path length curves 57, 58. In the case of the tilt angle =10, these path length curves 57 to 59 do not intersect over the whole x-dimension of the reticle 7. Therefore, situations enabling unwanted interference, as described above in conjunction with FIG. 21, do not occur. Thus, tilting the pupil facet mirror 20 in accordance with FIG. 22 leads to the various illumination channels, via which each of the field points of the illumination field 5 can be impinged upon by the partial beams of the illumination light 16, in each case having different channel lengths between the light source 2 and respectively one of the field points, i.e. respectively one of the points on the reticle 7 when the reticle is arranged for projection exposure.

(137) In accordance with FIG. 26, FIGS. 23 to 25 show the path length dependencies for the sixteen pupil facets discussed above in conjunction with FIGS. 22 and 26 on the x-dimension of the reticle 7. In the case of a tilt angle =7, the path length curves 57 to 59 likewise do not intersect, and so an arrangement of the pupil facet mirror 20 avoiding disturbing interference is also present here. In the case of a tilt angle of 4, the path length curves 58 and 59 intersect for those pupil facets 28 in the region of the pupil facet 28.sub.3 depicted outside right in FIGS. 20 and 22. FIG. 23 shows the situation in the case of a non-tilted pupil facet mirror 20 (=0), which corresponds to the situation according to FIGS. 20 and 21.

(138) A further measure, which can be used alternatively or additionally, of arranging the pupil facet mirror 20 in such a way that there is no unwanted identical optical path length opd of the illumination channels for the illumination light partial beams lies in the provision of a defined curvature of the main mirror surface 55 of the pupil facet mirror 20. This measure will be explained below on the basis of FIGS. 19, 27 and 28. In this variant, a curved main mirror surface 55 is present, which is indicated by a dash-dotted line in FIG. 19. FIG. 27 shows, with a magnified scale, a curvature profile 60 of the main mirror surface 55 of the pupil facet mirror 20 in the xz-plane (y=constant). Here, the left-hand scale in the diagram in FIG. 27 applies. At the same time, the diagram in FIG. 27 shows a curvature profile 61 of the curved main mirror surface 55 with a scale which is identical for the horizontal (x) and vertical (z) axes. In this case, the right-hand scale of the diagram in FIG. 27 applies.

(139) The main mirror surface 55 has a parabolic curvature profile.

(140) FIG. 28 shows the dependence of the optical path length opd for fifteen pupil facets 28, aligned equidistantly between the positions of the pupil facets 28.sub.1 and 28.sub.3 of the arrangement according to FIG. 20, emerging from this curved profile of the main mirror surface 55, i.e. the field profile (profile curves 62) of the optical path lengths of these fifteen pupil facets 28.

(141) The main mirror surface 55 has such a curvature that the fifteen pupil facets 28 practically have the same distance from the left-hand edge of the reticle 7 (x=52 mm). A small difference in the distance, which is too small to be able to be identified in the figure, is also present here and prevents disturbing equality of the optical path lengths to the left-hand edge of the reticle 7. The profile curves 62 do not intersect between x=52 and x=+52 mm, i.e. over the whole used region of the reticle 7. Thus, what emerges here is a situation in which disturbing equality of the optical path lengths from the various pupil facets 28 to specific points on the reticle 7 is also avoided once again.

(142) Thus, the main mirror surface 55 of the pupil facet mirror 20 deviates from a flat surface (reference surface 55) in such a way that the various illumination channels, via which each of the field points of the illumination field 5 can be impinged upon by partial beams of illumination light 16, in each case have different channel lengths between the light source 2 and respectively one of the field points of the illumination field 5.

(143) A further embodiment for avoiding disturbing interferences between illumination light partial beams guided over various illumination channels to the illumination field 11 is explained below on the basis of FIGS. 29 to 31. These show, with great magnification, respectively one of the field facets 25, this time with an arcuate or partial ring-shaped embodiment. A different scale is used for the horizontal axis and for the vertical axis in the drawing. The y-extent of the field facet 25 is one tenth of the x-extent thereof.

(144) The field facet 25 according to FIG. 29 is optimized for an operation of the projection exposure apparatus 1 with an illumination setting, i.e. a distribution of the illumination angles with which the object field points are illuminated, having the y-dipole configuration. In this y-dipole illumination setting, the illumination or object field 5 is illuminated from the direction of two illumination light poles spaced apart from one another in the y-direction. In this illumination setting, there is an intensity distribution in a pupil plane of the illumination optical unit 4 or 26 in the form of two intensity poles spaced apart from one another in the y-direction. Examples of such and further illumination settings are discussed in DE 10 2008 021 833 B4.

(145) The field facet 25 according to FIG. 29 carries blocking regions 63 on the reflection surface 64 thereof. The blocking regions 63 are embodied in a line-shaped manner. The blocking regions 63 block illumination light 16 incident on the reflection surface 64 of the field facet 25 such that the illumination light does not contribute to the illumination of the illumination field 5.

(146) The blocking regions 63, perpendicular to the longitudinal extent thereof, have a very small extent in the region of e.g. 100 m or even less. The blocking regions 63 have been applied to the field facet 25 where illumination light, which is guided through the illumination channel to which this field facet 25 belongs, for a specific object field point via this field facet 25 would have disturbing interference with illumination light from a further illumination channel to which a different field facet 25 belongs. Perpendicular to the respective longitudinal extent of the blocking regions 63, these have an extent which, for example, is at most 1 mm and is regularly significantly smaller, e.g. 800 m, 700 m, 600 m, 500 m, 400 m or may be even smaller still. An extent of the blocking regions 63 perpendicular to the direction of extent thereof which is greater than 1 mm is also possible. Hence, all blocking regions 63 of a field facet 25 together block less than 10% of the overall reflection surface 64 of this field facet 25.

(147) The blocking regions 63 block a corresponding portion of the cross section of the illumination light partial beam 16, incident on the field facet 25.

(148) Thus, what the blocking regions 63 bring about is that a cross-sectional region of at least one of the illumination channels of an illumination channel pair does not, due to the blocking regions 63, contribute to illuminating the illumination field 5 in the case of illumination channel pairs in which, during operation of the illumination optical unit 4 or 26, the corresponding pair of illumination light partial beams 16, which are guided via the illumination channels of the illumination channel pair, have a travel time difference, as calculated from the source, at any time which is shorter than the coherence duration .sub.K of the illumination light 16 at any respectively considered point of the illumination field 5. Therefore, disturbing interference components are suppressed by the blocking regions 63.

(149) In the case of field facets 25, appropriately equipped with such blocking regions 63, of the whole field facet mirror 19, an individual pattern of blocking regions 63 emerges on the reflection surface 64 for each of the field facets 25, depending on the arrangement thereof on the field facet mirror 19, depending on the geometry of the illumination optical unit 4 and depending on the illumination setting. By way of example, this individual pattern can be calculated by a ray tracing program. Depending on the result of this calculation, the pattern can then be applied to the reflection surfaces 64 of the field facets 25, for example by an appropriate coating.

(150) An integral light loss due to the blocking regions 63, i.e. a throughput loss for the illumination light 16, is less than 10% and may lie in the region of 2%.

(151) In principle, the field facet mirror 25 may also be built up from a plurality of individual micromirrors 65. A corresponding row-by-row and column-by-column array structure of such micromirrors 65 is indicated in FIG. 29 in sections. In such an alternative micromirror configuration, micromirrors 65, which are tilted in such a way that illumination light 16 impinging thereon does not contribute to the illumination of the object field 5, are present instead of the blocking regions 63.

(152) The micromirrors 65 are components of a MEMS array arrangement. By way of example, a corresponding MEMS array arrangement is known to a person skilled in the art from e.g. WO 2009/100 856 A1. The micromirrors 65 can be tilted independently of one another by at least two tilt degrees of freedom such that a reflecting guide of the illumination light partial beam 16 incident on such a micromirror field facet 25 can be predefined by the tilt position of the micromirrors 65.

(153) FIG. 30 shows a further field facet 25 in an illustration similar to FIG. 29. The field facet 25 according to FIG. 30 likewise carries blocking regions 63. In the field facet 25 according to FIG. 30, the arrangement of the blocking regions 63 is optimized for an x-dipole illumination setting. This arrangement of the blocking regions 63 differs qualitatively and quantitatively from the one according to FIG. 29.

(154) FIG. 31 shows a further embodiment of a field facet 25, once again in an illustration corresponding to the one according to FIGS. 29 and 30.

(155) In the case of the field facet 25 according to FIG. 31, blocking regions 63 have been applied for a plurality of different illumination settings that can be used in a projection exposure apparatus 1. These include the y-dipole and x-dipole illumination settings, and so all blocking regions 63 in FIGS. 29 and 30 are also present in the field facet 25 according to FIG. 31.

(156) In the case of a micromirror embodiment of the field facet 25, the various blocking region configurations according to FIGS. 29 to 31 can be set by appropriate tilt actuations of the micromirrors 65.

(157) A further embodiment of an illumination optical unit 4 for avoiding disturbing interferences of the illumination light partial beams 16, in the object field 5 is described on the basis of FIGS. 32 to 34. Components and functions corresponding to those which were already explained above with reference to FIGS. 1 to 31 are denoted by the same reference signs and will not once again be discussed in detail.

(158) The field facet mirror 19 according to FIGS. 32 to 34 once again has arcuate field facets 25, which are arranged next one another in four columns 25.sub.I, 25.sub.II, 25.sub.III and 25.sub.IV.

(159) The field facet mirror 19 simultaneously constitutes the optical retardation component 29 or 33 of the illumination optical unit 4.

(160) The facet columns 25.sub.I to 25.sub.IV are arranged offset to one another in the z-direction in relation to a main reflection plane 66 of the field facet mirror, i.e. perpendicular to the main reflection plane 66. Here, as indicated in FIG. 33, the offset of adjacent facet columns 25.sub.I to 25.sub.IV is l/2, i.e. it is at least half the coherence duration .sub.. What was already explained above, in particular in conjunction with the embodiment according to FIG. 4, applies to this path length difference. Alternatively, a step distance between the field facet columns 25.sub.I to 25.sub.IV may also be l/2. In this case, what was already explained above in conjunction with the embodiment according to FIG. 7 applies here.

(161) The field facet mirror is simultaneously illuminated by the illumination light 16 from the EUV light source 2 via a scanning region 67, as was explained above, for example in conjunction with the scanning regions 46 and 48 of the embodiments according to FIGS. 14 to 17. In the x-direction, the scanning region 67 has an extent corresponding to the x-extent of the whole field facet mirror 19. In the y-direction, the scanning region 67 has an extent corresponding to the y-extent of respectively one of the field facets 25.

(162) What the stepped arrangement of the facet columns 25.sub.I to 25.sub.IV ensures is that illumination light partial beams 16.sub.i, which are guided to different field facet columns 25.sub.I to 25.sub.IV, do not have disturbing interference.

(163) FIG. 34 shows, in an exemplary manner, an assignment of four illumination channels 27 to the field facets 25, respectively the top ones in FIG. 34, of the four facet columns 25.sub.I to 25.sub.IV. In the very schematic illustration according to FIG. 34, the field facet mirror 19 and the pupil facet mirror 20 are respectively depicted in a top view. FIG. 34 only shows those pupil facets 28 of the pupil facet mirror 20 which belong to the four illumination channels 27 shown in an exemplary manner. The tilt angles of the field facets assigned to these four illumination channels 27 are such that a point-symmetric arrangement of pupil facets 28 is illuminated in the region of the arrangement plane of the pupil facet mirror 20. There moreover is such an assignment of the pupil facets 28 to the field facets 25 that the pupil facets 28, assigned to the field facets 25 of one of the field facet columns 25.sub.I to 25.sub.IV via respective illumination channels 27, are arranged symmetrically and uniformly distributed in the arrangement plane of the pupil facet mirror 20 in accordance with the respectively desired illumination setting. An unwanted negative influence of illumination parameters, in particular on a telecentricity and an ellipticity of the illumination, due to the z-offset of the field facet columns 25.sub.I to 25.sub.IV can then be reduced or completely avoided. Definitions of these telecentricity and ellipticity illumination parameters can be found in U.S. Pat. No. 8,174,677.

(164) The arrangement example of the illumination channels 27 according to FIG. 34 belongs to an annular illumination setting or a quadrupole illumination setting.

(165) Due to the larger scanning region 67 compared to the embodiments according to FIGS. 14 to 17, this results in a higher throughput of the illumination optical unit 4. Moreover, effects of light source variations on illumination variables can be reduced.