DE102014226921A1 - The radiation source module - Google Patents

Abstract

A radiation source module (2) for an illumination system (19) of a projection exposure system (1) having a plurality M of scanners (3i) comprises a coupling-out optical system (9) having at least one means for dividing a collective output beam (8) into a plurality of total intensity (M) of single output beams (10i) having single intensities, wherein the means for dividing the collection output beam (8) comprises means for varying a division ratio of the total intensity of the collection output beam (8) into the single intensities of the single output beams (10i) includes.

Description

The invention relates to a radiation source module, in particular a radiation source module for an illumination system of a projection exposure system with a plurality of scanners. The invention further relates to a lighting system for a projection exposure system with a plurality of scanners. In addition, the invention relates to a method for controlling the distribution of illumination radiation of a common radiation source to a plurality of scanners. Finally, the invention relates to a projection exposure system for projection lithography, to a method for producing a microstructured or nanostructured component, and to a component produced by this method.

The illumination radiation used in the projection exposure to image a pattern on a wafer may be subject to variations. Such variations can cause the wafer to be exposed unevenly.

It is an object of the invention to improve a radiation source module, in particular a radiation source module for an illumination system of a projection exposure system with a plurality of scanners. In particular, it is an object of the invention to improve a coupling-out optical system for dividing a common collective output beam into scanner-specific single output beams. The coupling-out optical system forms in particular a component of the radiation source module.

This object is achieved by a coupling-out optical system in which the means for dividing the collective output beam into single output beams comprises means for varying a distribution ratio of the total intensity of the collective output beam into the individual intensities of the single output beams.

In this case, the optical device which serves to split a raw beam emitted from a common radiation source unit with illumination radiation, if appropriate after it has been formed into a collective output beam, into single output jets is referred to as a coupling-out optical system, each of the individual output jets being exposed to a specific scanner, in particular one specific beam guiding optics, is assigned. Different individual output beams each serve to illuminate different object fields, in particular spatially separate object fields.

The collecting output beam can be divided into at least two, in particular at least three, in particular at least five, in particular at least eight, in particular at least ten individual output jets. The number of individual output beams corresponds in particular to the number of scanners of the projection exposure system. This number is at most 30.

According to the invention, it has been recognized that the stability of the scan-integrated intensity of the illumination of the individual object fields and thus the dose stability for illuminating the wafers can be improved by varying the distribution ratio of the total intensity of the collection output beam in individual intensities of the individual output beams.

In particular, the coupling-out optical system comprises a means for temporally varying the distribution ratio of the total intensity of the collective output beam into the individual intensities of the single output beams. The distribution ratio can be varied in a targeted manner in particular. It can also be varied periodically. According to the invention, it has been recognized that the dose stability during exposure of the wafer depends less on the stability of the instantaneous illumination of the wafer, but rather on the stability of the scan-integrated overall intensity during the exposure of the wafer. By a variation, in particular a temporal variation of the distribution ratio of the total intensity into the individual intensities, the overall dose stability can be substantially improved.

A variation of the distribution ratio of the total intensity can be achieved by targeted attenuation of one or more, in particular a targeted selection, of the single output beams. A variation of the division ratio of the total intensity may also include a redistribution of the intensities of the single output beams, that is, a part of the single intensity of one of the single output beams is selectively supplied to another of the single output beams.

The variation can in particular be timed.

According to one aspect of the invention, the means for dividing the collecting output beam into single output jets comprises at least one beam guiding element, which is designed such that it is in Dependence of its relative position to the collection output beam results in divisions of the collection output beam into single output beams, in which the distribution of the total intensity of the collection output beam to the single output beams is different. The beam guiding element is in particular displaceable, in particular actuatable displaceable. It can also be displaced periodically. Instead of the beam guiding element, the collecting output beam can also be displaceable. In particular, it is provided that the impact position of the output beam on the beam guidance element is variable, in particular controllably displaceable, in the form of a train. With the aid of such a beam-guiding element, a targeted variation of the distribution ratio of the total intensity of the collective output beam into the individual intensities of the single output beams is possible in a simple manner. Such a beam-guiding element also allows easy control of the variation of the distribution ratio of the total intensity to the individual intensities.

The beam-guiding element is also called a separation component. The separation component comprises in particular at least one radiation-reflecting element. It also comprises at least one displaceable, in particular at least one actuatable displaceable element. The at least one displaceable element may be identical to the at least one radiation-reflecting element. They can also be different, separate elements.

The at least one displaceable element can be displaced by means of a piezoactuator. Piezo actuators allow a precise and very fast displacement. The displacement can take place faster than 1 ms, in particular faster than 100 μs, in particular faster than 10 μs. By means of a piezo actuator and an axis, a rotational displacement or tilting can be achieved. By means of two oppositely controlled piezo actuators tilting can be achieved. By means of an electromagnetic actuator, for example a plunger drive, a displacement can take place over great distances. The displacement can be carried out in particular by at least 100 .mu.m, in particular 1 mm, in particular 10 mm.

The actuator is preferably suitable for vacuum, that is, vacuum is not detrimental to its operation. Preferably, the actuator does not emanate, that is, it does not emit gaseous substances, at least no gases that can chemically react or be activated under EUV radiation. Preferably, the movement of the actuator and / or the component moved by the actuator does not lead to the release of small particles of a solid material, in particular no particles with sizes between 1 nm and 1 μm.

The radiation-reflecting element of the separation component is larger in one direction than the instantaneous illumination caused by the collection output beam on it. The collection output beam thus leads to the illumination of only a portion of the radiation-reflecting element of the separation component. In particular, the subregion makes up at most 50%, in particular at most 20%, in particular at most 10%, in particular at most 5%, of the entire reflection surface of the radiation-reflecting element of the separation component.

The portion of the radiation-reflecting element which is illuminated by the collection output beam, in turn, comprises different radiation-reflecting regions which are assigned to the different scanners, that is, which lead to the transfer of certain of the individual output beams to specific scanners.

Different partial regions of the radiation-reflecting element of the separation component differ, for example, by the relative size of the different regions assigned to the different scanners. Therefore, by shifting the collection output beam relative to the radiation-reflective element of the separation component, the distribution ratio of the collection output beam to the single output beams can be easily varied. Details and details emerge from the description of different embodiments.

According to one aspect of the invention, the radiation source module comprises means for periodically varying the division ratio of the total intensity of the collection output beam into the individual intensities of the single output beams. In particular, the variation of the division ratio has a period T _Var which is at most 10 ms, in particular at most 5 ms, in particular at most 3 ms, in particular at most 2 ms, in particular at most 1 ms, in particular at most 0.5 ms, in particular at most 0.3 ms , in particular at most 0.2 ms, in particular at most 0.1 ms, in particular at most 0.05 ms, in particular at most 0.03 ms, in particular at most 0.02 ms, in particular at most 0.01 ms.

The period T _Var of the distribution ratio variation is, in particular, shorter than the time T _Scan , which requires a point on the wafer to pass through the scan slot. In particular: T _Var : T _Scan ≦ 1, in particular T _Var : T _Scan ≦ 0.5, in particular T _Var : T _Scan ≦ 0.3, in particular T _Var : T _Scan ≦ 0.3, in particular T _Var : T _Scan ≤ 0.2, in particular T _Var : T _Scan ≤ 0.1, in particular T _Var : T _Scan ≤ 0.05, in particular T _Var : T _Scan ≤ 0.03, in particular T _Var : T _Scan ≤ 0.02, in particular T _Var : T _Scan ≤ 0.01.

This ensures that the scan-integrated dose can be kept constant to illuminate any point on the wafer.

Preferably, the time T _scan , which takes a point on the wafer to pass through the scan slot, is just an integer multiple of the period T _Var . Alternatively and / or in addition thereto, the _{scan time} T _{scan is} substantially greater than the period T _Var , in particular T _Scan : T _Var ≥ 100, in particular T _Scan : T _Var ≥ 1000. This also ensures that each area on the wafer substantially is illuminated with a dose averaged over the entire period of variation of the division ratio.

According to a further aspect of the invention, the radiation source module comprises at least one means for controlling the overall intensity of the collection output beam.

By controlling the total intensity, the absolute radiation dose for exposure of the wafer can be controlled. The control of the total intensity of the collection output beam is possible very quickly. It can be controlled in particular with a frequency of at least 10 kHz, in particular at least 100 kHz, in particular at least 1 MHz, in particular at least 10 MHz, in particular at least 100 MHz, in particular 1 GHz. The absolute radiation dose is precisely adjustable. In particular, it can be controlled with an accuracy of better than 1%, in particular better than 1 ‰, in particular better than 0.1 ‰. Such high accuracy can be achieved in particular by means of a feedback loop.

The control of the total intensity of the collecting output beam can be achieved with the aid of a beam shaping optical system, which serves in particular for shaping the collecting output beam from a raw beam with illumination radiation emitted by the radiation source unit.

According to one aspect of the invention, the radiation source itself, in particular for generating the raw beam, controllable. The radiation source, which may also be a radiation source unit with one or more radiation sources, serves in particular for generating a raw beam having a controllable total intensity I _{SQ of} the radiation source, the total intensity I _SQ of the raw beam emitted by the radiation source having a frequency of at least 10 kHz, in particular at least 100 kHz, in particular at least 1 MHz, in particular at least 10 MHz, in particular at least 100 MHz, in particular 1 GHz is controllable.

In particular, the radiation source is an EUV radiation source, that is to say a radiation source which has illumination radiation in the EUV range, in particular in the wavelength range from 2 nm to 30 nm, in particular in the wavelength range from 2 nm to 15 nm, in particular illumination radiation with a Wavelength of 13.5 nm or 6.7 nm, emitted.

In particular, the radiation source may be a free electron laser (FEL) or a synchrotron-based radiation source. Such radiation sources are particularly advantageous for illumination systems for projection exposure systems with a plurality of scanners. They are particularly suitable for generating a collection output beam having a high overall intensity. In particular, they are very quickly controllable.

According to one aspect of the invention, the total intensity I _{SQ of} the illumination radiation emitted by the radiation source unit, in particular by the radiation source, can be controlled with a frequency f _S which is at least as great as the product of the number M of individual output beams and the reciprocal of the period T _{Var of} the variation of the distribution ratio of the total intensity of the collection output beam into the single output beams, f _S ≥ M × 1 / T _var .

This is possible in a particularly simple manner if the radiation source emits radiation pulses having at least one predetermined pulse frequency. The pulse frequency f _{pulse of} the radiation source is in particular in the range of 100 kHz to 10 GHz. It is in particular at least 1 MHz, in particular at least 10 MHz.

With the designation f _Var = 1 / T _{Var, the} following applies in particular: f _pulse : f _Var ≥ M, in particular f _pulse : f _Var ≥ 10 M, in particular f _pulse : f _Var ≥ M, in particular f _pulse : f _Var ≥ 1000 M, in particular f _pulse : f _Var ≥ 10 ⁴ M, in particular f _pulse : f _Var ≥ 10 ⁵ M, in particular f _pulse : f _Var ≥ 10 ⁶ M.

According to a further aspect of the invention, the radiation source module comprises beam shaping optics for shaping the collection output beam from the raw beam with illumination radiation emitted by the radiation source unit, in particular the radiation source. With the aid of the beam shaping optical system, a collective output beam with a specific intensity profile, in particular a collective output beam with a specific cross section and / or a specific divergence, can be generated from the raw beam. The collecting output beam can in particular be shaped so that it has an elongated, in particular a strip-shaped, in particular a rectangular cross-section whose longer side corresponds to just one side length of the separation component, while in the direction perpendicular thereto has significantly smaller dimensions than the separation component in the corresponding direction. In particular, the shorter side is at most 0.5 times as long, in particular at most 0.3 times as long, in particular at most 0.1 times as long, in particular at most 1 / M times as long, as the corresponding direction of the separation component ,

The beam shaping optics may include one or more means for influencing the intensity of the collection output beam, in particular for attenuating the collection output beam.

According to a further aspect of the invention, the means for controlling the total intensity of the collecting output beam in dependence on the means for dividing the collecting output beam into individual output jets is controllable, in particular controllable.

According to the invention, it has been recognized that it is advantageous for controlling the radiation dose, which is used for exposure of the wafer, in particular for ensuring the dose stability, the relative dose of the single output jets, that is the ratio of the radiation doses in different scanners, and the absolute radiation dose to steer separately. The control of the absolute radiation dose can be achieved in a simple manner by controlling the overall intensity of the collection output beam. In particular, the total intensity can be controlled in time as a function of the different divisions of the collective output beam into single output beams. In particular, it is possible to increase or decrease the overall intensity of the collection output beam during certain intervals of the period T _Var . In this way, it can be achieved that each of the scanners receives a predetermined radiation dose for the exposure of the wafer averaged over time, that is to say averaged over a period T _Var .

The radiation dose for exposure of the wafer is directly related to the single intensity of the respective single output beam. The control of the radiation dose thus corresponds directly to the control of the respective intensity.

The intensity of the illumination radiation emitted by the radiation source is directly proportional to the pulse frequency of the radiation pulses emitted by the radiation source at a constant pulse energy. The intensity of the illumination radiation emitted by the radiation source can be controlled particularly easily via the pulse frequency of the radiation source.

The means for controlling the overall intensity of the collective output beam can be arranged in the beam path in front of or behind the beam-shaping optical system. In particular, it is possible to control the overall intensity of the collective output beam via the pulse frequency of the radiation source, that is to say via the pulse frequency of the emitted raw beam.

It is also possible to control the intensity of the collective output beam after or during its formation in the beam-shaping optical system, in particular to reduce it.

According to a further aspect of the invention, the means for controlling the overall intensity of the collection output beam is dependent on a means for specifying setpoint output. Intensities of the individual output beams controllable, in particular controllable. The total intensity I _SQ of the raw beam emitted by the radiation source and / or of the collective output beam can be controlled, in particular regulated, in particular as a function of the sum of all of the individual desired intensities.

According to a further aspect of the invention, the means for varying the division ratio of the total intensity into the individual intensities in dependence on a predetermined ratio of the desired intensities of the individual output jets is controllable, in particular controllable. For controlling the overall intensity of the raw beam and / or the collection output beam and / or for controlling the means for varying the division ratio, a separate control unit, in particular a computer-aided control unit, may be provided.

According to another aspect of the invention, at least 50% of the single output jets have at least 50% of the period T _Var individual intensities, which is at least 10% of the average single intensity of all the single output jets.

According to a further aspect of the invention, a proportion of at least 50%, in particular at least 70%, in particular at least 90%, in particular 100%, of the individual output beams has a variance, determined over the period T _Var , which is smaller than a quarter of the squared average Single intensity of the respective single output beam, determined over the period T _Var , is.

Another object of the invention is to improve a lighting system for a projection exposure system with a plurality of scanners.

This object is achieved by an illumination system with a radiation source module according to the preceding description and a plurality M of beam guiding optics for guiding the individual output beams to different object fields, in particular to spatially separate object fields.

The advantages result from those of the radiation source module.

Another object of the invention is to improve a method of controlling the distribution of illumination radiation from a common radiation source to a plurality of scanners.

• – Bereitstellen eines Beleuchtungssystems gemäß der vorhergehenden Beschreibung,
• – Vorgeben von Soll-Intensitäten der Einzel-Ausgabestrahlen,
• – Bestimmen eines zeitlichen Verlaufs des Aufteilungsverhältnisses der Gesamtintensität des Sammel-Ausgabestrahls in die Einzelintensitäten der Einzel-Ausgabestrahlen in Abhängigkeit einer periodisch veränderlichen Einstellung des Mittels zur Variation dieses Aufteilungsverhältnisses,
• – Bestimmen der mittleren Ist-Intensitäten jedes der Einzel-Ausgabestrahlen, insbesondere Bestimmen der mittleren Ist-Intensitäten am Eingang jedes der Scanner und/oder Bestimmen der mittleren Ist-Intensitäten, mit welchen die unterschiedlichen Objektfelder beleuchtet werden, wobei unter der mittleren Ist-Intensität jeweils die über eine Periode T_Var gemittelte Intensität verstanden sein soll,
• – Bestimmen von Korrekturfaktoren zur Anpassung der mittleren Ist-Intensitäten jedes der Einzel-Ausgabestrahlen an die vorgegebenen Soll-Intensitäten und
• – Steuern der Gesamtintensität I_SQ des Sammel-Ausgabestrahls in Abhängigkeit von den Korrekturfaktoren und in Abhängigkeit von der periodischen Veränderung des Mittels zur Variation des Aufteilungsverhältnisses. Hierbei kann die Gesamtintensität I_SQ des Sammel-Ausgabestrahls insbesondere durch Steuerung der Strahlungsquelle, insbesondere durch Steuerung der Pulsfrequenz der Strahlungsquelle, erreicht werden.

This task is solved by a procedure with the following steps:

• Providing a lighting system according to the preceding description,
• Predetermining target intensities of the single output jets,
• Determining a time profile of the distribution ratio of the total intensity of the collective output beam into the individual intensities of the single output beams as a function of a periodically variable setting of the means for varying this distribution ratio,
• Determining the average actual intensities of each of the individual output beams, in particular determining the average actual intensities at the input of each of the scanners and / or determining the mean actual intensities with which the different object fields are illuminated, below the mean actual intensity in each case the intensity averaged over a period T _Var should be understood,
• Determining correction factors for adjusting the average actual intensities of each of the individual output beams to the predetermined desired intensities and
• - controlling the total intensity I _SQ of the collecting output beam in response to the correction factors and in response to the periodic variation of the means for varying the division ratio. In this case, the total intensity I _{SQ of} the collective output beam can be achieved in particular by controlling the radiation source, in particular by controlling the pulse frequency of the radiation source.

The total intensity of the collecting output beam, in particular the total intensity of the raw beam emitted by the radiation source, can be regulated in particular, in particular regulated continuously.

The total intensity of the collection output beam can be controlled in particular periodically. In particular, it can be time-controlled periodically with the period T _Var and / or within the period T _Var . In particular, it is possible to divide the period T _Var into a sequence of disjoint time intervals in which the total intensity I _{SQ of} the collection output beam is constant. The total intensity I _{SQ of} the collection output beam may be different in the different subintervals of the period T _Var .

The object is also achieved by a method in which the actual intensities of each of the single output jets are determined in dependence on the variable setting of the divider ratio varying means, then one or more settings of the divisional variation means are determined in order to achieve a desired distribution ratio of the total intensity of the collection output beam to the single output beams, and then at least one of these settings, in particular a sequence of these settings, is set. The latter can be controlled, in particular regulated happen.

In this method too it can be provided to control the total intensity I _{SQ of} the collective output beam, in particular depending on the setting of the means for varying the distribution ratio of the total intensity of the collective output beam into the individual intensities of the single output beams.

Another object of the invention is to improve a projection exposure system for projection lithography.

This object is achieved by a projection exposure system with a lighting system according to the preceding description and a plurality M of scanners. The advantages result from those of the lighting system.

The projection exposure system according to the invention makes it possible in particular to stably supply a plurality of scanners with a single radiation source, in particular a FEL, with illumination radiation. In particular, the projection exposure system according to the invention leads to improved dose stability in the exposure of wafers in a plurality of independent, separate scanners, which are supplied with illumination radiation by a common radiation source unit.

Further objects of the invention are to improve a method for producing micro- or nanostructured components as well as components produced according to the method. These objects are achieved by providing a projection exposure system as described above. The advantages result from the previously described.

Further details and advantages of the invention will become apparent from the description of embodiments with reference to the drawings. Show it:

1 a schematic representation of the components of a projection exposure system with a plurality of scanners,

2 an alternative schematic representation of the projection exposure system with a plurality of scanners,

3 another alternative of the projection exposure system,

4 1 is a further schematic representation of the projection exposure system with details of a radiation source module according to a first variant,

5 a representation for explaining details of the radiation source module with an additional imaging optics,

6 an enlarged detail of the separation plate of the radiation source module according to 5 .

7A to 7D exemplary representations of alternative variants of the separation plate,

8th to 10 other alternatives of the separation plate,

11 to 16 other alternatives of the separation plate as grid representations,

17 exemplary representations of the efficiency of the distribution of the collective output beam into single output jets as a function of the number of scanners for different setting ranges r,

18 to 23 exemplary representations of the energy loss as a function of a maximum redistribution u of radiation intensity between the individual single output jets for a different number M of scanners,

24 a representation according to 4 an alternative variant of the coupling-out optics,

25 and 26 Details from the 24 .

27 a representation according to 26 with an alternative arrangement of the different regions of the separation plate,

28 and 29 Representations according to 4 with alternative variants of the coupling-out optics,

30 a representation according to 5 a further variant of the coupling-out optics,

31 a further variant of the coupling-out optics according to 3 .

32 to 34B schematic diagrams for explaining the method for adjusting the power of the radiation source as a function of the distribution of the collective output beam on the individual output beams by means of the coupling-out,

35 and 36 schematic representations to illustrate further aspects of the method for dividing the collection output beam onto single output jets,

37 and 38 schematic representations of different variants of the separation plate to illustrate further aspects of the invention,

39 and 40 schematic representations of different variants of the functions which describe the distribution ratio of the collective output beam into single output beams over the period T _Var ,

41 and 42 schematic representations of optical components for the parallel displacement of an illumination beam,

43 a schematic representation of an angle magnification components, and

44A to D schematic representations of the individual functions, each of which is the deviation of the area of the areas of the separation plates according to the 7A to 7B represent a nominal value.

The following are with reference to the 1 First, the essential components of a projection exposure system 1 described.

Subsequent division of the projection exposure system 1 subsystems primarily serve to delineate them. The subsystems can form separate structural subsystems. However, the division into subsystems does not necessarily have to be reflected in a constructive demarcation.

The projection exposure system 1 includes a radiation source module 2 and a plurality of scanners 3 _i .

The radiation source module 2 includes a radiation source 4 for generating illumination radiation 5 ,

At the radiation source 4 it is in particular a free electron laser (FEL). It can also be a synchrotron radiation source or a synchrotron radiation-based radiation source which generates coherent radiation with very high brilliance. As an example for such radiation sources on the US 2007/0152171 A1 and the DE 103 58 225 B3 directed.

The radiation source 4 For example, it has an average power in the range of 1 kW to 35 kW. It has a pulse rate in the range of 10 MHz to 10 GHz. Each individual radiation pulse may for example have an energy of 83 μJ. With a radiation pulse length of 100 fs, this corresponds to a radiation pulse power of 833 MW.

The radiation source 4 can also have a repetition rate in the kilohertz range, for example, of 100 kHz, or in the low megahertz range, for example at 3 MHz, in the middle megahertz range, for example at 30 MHz, in the upper megahertz range, for example at 300 MHz or in the gigahertz range, for example at 1.3 GHz , own.

At the radiation source 4 it is in particular an EUV radiation source. The radiation source 4 in particular emits EUV radiation in the wavelength range, for example between 2 nm and 30 nm, in particular between 2 nm and 15 nm.

The radiation source 4 emits the illumination radiation 5 in the form of a raw jet 6 , The raw beam 6 has a very small divergence. The divergence of the raw jet 6 may be less than 10 mrad, in particular less than 1 mrad, in particular less than 100 urad, in particular less than 20 urad. To simplify the description of positional relationships, coordinates of a Cartesian xyz coordinate system are used below. The x-coordinate regularly tightens a bundle cross-section of the illumination radiation with the y-coordinate 5 on. The z-direction runs regularly in the radiation direction of the illumination radiation 5 , In the area of the object plane 21 or the picture plane 24 the y-direction is parallel to a scan direction. The x-direction is perpendicular to the scan direction. The raw beam 6 is from the radiation source 4 emitted in a certain direction. The deviation of this direction from a given nominal direction is also referred to below as pointing.

The raw beam 6 may have an optical conductivity which is less than 0.1 mm ² , in particular less than 0.01 mm ² . The optical conductivity is the smallest volume of a phase space, which is 90% of the energy of the radiation source 2 emitted illumination radiation 5 contains. Corresponding definitions of the light conductance can be found, for example, in US Pat EP 1 072 957 A2 and the US 6,198,793 B1 ,

The radiation source module 2 further comprises one of the radiation source 4 downstream beamforming optics 7 , The beam shaping optics 7 serves to generate a collection output beam 8th from the raw stream 6 , The collection output beam 8th has a very small divergence. The divergence of the collection output beam 8th may be less than 10 mrad, in particular less than 1 mrad, in particular less than 100 urad, in particular less than 10 urad.

By means of the beam shaping optics 7 in particular, the diameter of the raw jet 6 or the collection output beam 8th to be influenced. By means of the beam shaping optics 7 In particular, a widening of the raw jet 6 be achieved. The raw beam 6 can by means of the beam shaping optics 7 in particular by a factor of at least 1.5, in particular at least 2, in particular at least 3, in particular at least 5, in particular at least 10 are widened. In particular, the expansion factor is less than 1000. It is also possible to use the raw beam 6 expand differently in different directions. In particular, it can be widened more in an x-direction than in a y-direction. Here, the y-direction corresponds to the area of the object field 11 _i the scanning direction. The divergence of the collection output beam 8th may be less than the divergence, especially less than half the divergence, of the raw beam 6 ,

For further details of the beam shaping optics 7 be on the DE 10 2013 223 935.1 referred to, which is hereby incorporated into the present application. The beam shaping optics 7 may in particular have one or two beam-forming mirror groups each having two mirrors. The beam-forming mirror groups are used in particular for beam shaping of the collective output beam 8th in mutually perpendicular planes which are parallel to the propagation direction of the collection output beam 8th run.

The beam shaping optics 7 may also include other beam shaping mirrors.

The beam shaping optics 7 may in particular cylinder mirror, in particular at least one convex and at least one concave cylindrical mirror include. It can also include mirrors with a free-form profile. Such mirrors each have a height profile which can not be represented as a conic section.

By means of the beam shaping optics 7 can also the intensity profile of the raw beam 6 to be influenced.

In addition, the radiation source module comprises 2 one of the beam shaping optics 7 downstream, described in more detail below, coupling-out 9 , The decoupling optics 9 serves to generate several, namely n, single output jets 10 _i (i = 1 to n) from the collection output beam 8th , The single output jets 10 _i each form beam bundles for illuminating an object field 11 _i . The single output jets 10 _i are each one of the scanners 3 _i assigned. The beams of the single output jets 10 _i can each have a plurality of separate partial beams 12 _i include.

The radiation source module 2 is arranged in particular in an evacuable housing.

The scanners 3 _i each comprise a beam guiding optics 13 _i and a projection optics 14 _i .

The beam guiding optics 13 _i serves to guide the illumination radiation 5 , in particular the respective individual output beams 10 _i to the object fields 11 _i the single scanner 3 _i .

The projection optics 14 _i is used in each case to represent one in one of the object fields 11 _i arranged Retikels 22 _i in an image field 23 _i , in particular one in the image field 23 _i arranged wafer 25 _i .

The radiation guidance optics 13 _i comprises in the order of the beam path of the illumination radiation 5 each a deflection optics 15 _i, coupling optics 16 _i , in particular in the form of a focusing assembly, and an illumination optics 17 _i . The coupling optics 16 _In particular, _i can also be designed as a Wolter type III collector.

The deflection optics 15 _i can also be used in the coupling-out optics 9 be integrated. The decoupling optics 9 may in particular be designed such that they the individual output jets 10 _i already redirects in a desired direction. According to a variant can also on the deflection optics 15 _{i be} waived altogether. In general, the coupling-out optics 9 and the deflection optics 15 _{i form} a decoupling deflecting device.

For different variants of the deflection optics 15 _i am for example on the DE 10 2013 223 935.1 referenced, which is hereby incorporated as part of the present application in this.

The coupling optics 16 _i is used in particular for coupling the illumination radiation 5 , in particular one of the coupling-out optics 9 generated single output jets 10 _i in each case one of the illumination optics 17 _i .

The beam guiding optics 13 _i forms together with the beam shaping optics 7 and the decoupling optics 9 Components of a lighting device 18 ,

The lighting device 18 is as well as the radiation source 4 Part of a lighting system 19 ,

Each of the lighting optics 17 _i is one of the projection optics 14 _i assigned. Together, the associated lighting optics 17 _i and the projection optics 14 _i also as optical system 20 _i denotes.

The illumination optics 17 _i is used in each case for the transfer of illumination radiation 5 to one in the object field 11 _i in an object plane 21 arranged reticle 22 _i . The projection optics 14 _i is used to image the reticle 22 _i , in particular for imaging structures on the reticle 22 _i , on one in a picture box 23 _i in an image plane 24 arranged wafers 25 _i .

The projection exposure system 1 in particular comprises at least two, in particular at least three, in particular at least four, in particular at least five in particular at least six, in particular at least seven, in particular at least eight, in particular at least nine, in particular at least ten scanners 3 _i . The projection exposure system 1 can have up to twenty scanners 3 _i include.

The scanners 3 _i are from the common radiation source module 2 , in particular the common radiation source 4 , with illumination radiation 5 provided.

The projection exposure system 1 serves for the production of micro- or nanostructured components, in particular electronic semiconductor components.

The coupling optics 16 _i is in the beam path between the radiation source module 2 , in particular the coupling-out optics 9 , and each one of the illumination optics 17 _i arranged. It is designed in particular as a focusing assembly. It is used to transfer one of the individual output jets 10 _i in an intermediate focus 26 _i in an intermediate focus level 27 , The intermediate focus 26 _i can in the region of a passage opening of a housing of the optical system 20 _i or the scanner 3 _{i be} arranged. The housing is in particular evacuated.

The illumination optics 17 Each _i comprises a first facet mirror and a second facet mirror whose function corresponds in each case to those known from the prior art. The first facet mirror may in particular be a field facet mirror. The second facet mirror may in particular be a pupil facet mirror. However, the second facet mirror may also be spaced apart from a pupil plane of the illumination optics 17 _{i be} arranged. This general case is also called a specular reflector.

The facet mirrors each include a variety of facets. When operating the projection exposure system 1 Each of the first facets is assigned one of the second facets.

The mutually associated facets each form an illumination channel of the illumination radiation 5 for illuminating the object field 11 _i at a certain illumination angle.

The channel-wise assignment of the second facets to the first facets takes place as a function of a desired illumination, in particular of a predetermined illumination setting. The facets of the first facet mirror can be displaceable, in particular tiltable, in particular with two tilting degrees of freedom. The facets of the first facet mirror are in particular switchable between different positions. They are assigned in different switching positions different of the second facets. In each case, at least one switching position of the first facets may be provided, in which the illuminating radiation impinging on them 5 not to illuminate the object field 11 _i contributes. The facets of the first facet mirror can be designed as virtual facets. This is understood to mean that they are formed by a variable grouping of a plurality of individual mirrors, in particular a plurality of micromirrors. For details be on the WO 2009/100856 A1 referenced, which is hereby incorporated as part of the present application in this.

The facets of the second facet mirror can accordingly be designed as virtual facets. They can also be correspondingly displaceable, in particular tiltable, be formed.

The first facets are introduced into the object field via the second facet mirror and optionally via a subsequent transmission optics, not shown in the figures, which comprises, for example, three EUV mirrors 11 _i in the reticle or object plane 21 displayed.

The individual illumination channels lead to the illumination of the object field 11 _i with certain lighting angles. The totality of the illumination channels thus leads to an illumination angle distribution of the illumination of the object field 11 _i by the illumination optics 17 _i . The illumination angle distribution is also referred to as the illumination setting.

In a further embodiment of the illumination optics 17 _i , in particular with a suitable position of the entrance pupil of the projection optics 14 _i , can look at the mirrors of the transfer optics in front of the object field 11 _i also be dispensed with, which leads to a corresponding increase in transmission for the Nutzstrahlungsbündel.

The reticle 22 _i with for the illumination radiation 5 reflective structures is in the object plane 21 in the area of the object field 11 _i arranged. The reticle 22 _i is carried by a reticle holder. The reticle holder is controlled by a displacement device displaced.

The projection optics 14 _i forms the object field in each case 11 _i in the image field 23 _i in the picture plane 24 from. In this picture plane 24 is in the projection exposure of the wafer 25 _i arranged. The wafer 25 _i has a photosensitive coating formed during the projection exposure with the projection exposure system 1 is exposed. The wafer 25 _i is carried by a wafer holder. The wafer holder is controlled by a displacement device displaced.

The displacement device of the reticle holder and the displacement device of the wafer holder can be in signal connection with one another. They are especially synchronized. The reticle 22 _i and the wafer 25 _i are in particular synchronized with each other displaced.

In the following, an advantageous embodiment of the illumination system 19 described.

It was recognized that as the main source of radiation 4 Advantageously, a free electron laser (FEL) or a synchrotron-based radiation source can be used. A FEL scales very well, which means that it can be particularly economically operated if it is designed large enough to a plurality of scanners 3 _i with illumination radiation 5 to supply. In particular, the FEL may have up to eight, ten, twelve or even twenty scanners with illumination radiation 5 supply.

It can also be more than a radiation source 4 be provided.

A requirement for the projection exposure system 1 is that the radiation intensity that the individual reticles 22 _i , and in particular the radiation dose which the wafers 25 _i reached, can be regulated very accurately and very quickly. The radiation dose, which the wafers 25 _i achieved, should be kept as constant as possible in particular.

Fluctuations of the reticle 22 _i incident illumination radiation 5 , in particular the total intensity of the on the reticle 22 _i incident illumination radiation 5 and so on the wafers 25 _i radiation dose, may be due to intensity fluctuations of the main radiation source and / or to geometric variations, in particular to variations in the direction of the main radiation source 4 emitted raw beam 6 and / or variations in the cross-sectional profile, in particular in the region of the coupling-out optical system 9 to be due to the same. Variations in the cross-sectional profile may be due, in particular, to divergence variations in the radiation source 4 emitted raw beam 6 and / or the collection output beam 8th be due. Fluctuations of the reticle 22 _i incident illumination radiation 5 may in particular also to transmission fluctuations between the radiation source 4 and the reticle 22 _i , in particular to transmission fluctuations between the coupling-out optics 9 and the scanner 3 _{i be} due.

The following are details of the coupling optics 9 described in more detail.

The following are details of different embodiments of the radiation source module 2 , in particular the coupling-out optics 9 , described. The radiation source module 2 is in particular a component of the lighting system 19 , The lighting system 19 also includes a plurality M of beam guiding optics 13 _i for guiding the single output jets 10 _i to the different object fields 11 _i the different scanner 3 _i .

The decoupling optics 9 is in particular a component of the lighting device 18 , The lighting device 18 forms together with the radiation source 4 or generally with a radiation source unit, which basically also a plurality of different radiation sources 4 may include the lighting system 19 ,

In the 2 is again very schematically the basic structure of the projection exposure system 1 shown. In the projection exposure system 1 a plurality M becomes the scanner 3 _i from the radiation source module 2 , in particular from the single radiation source 4 , with illumination radiation 5 provided. As a radiation source 4 In particular, a single, powerful Free Electron Laser (FEL) is used. The total power of the illumination radiation 5 which of such a radiation source 4 is hardly in the range of 1 kW to 35 kW, in particular in the range of 10 kW to 25 kW.

According to the in 3 variant shown is each of the scanner 3 _i a sensor 32 _i assigned. At the sensor 32 _i is in particular an energy sensor. With the help of the sensors 32 _i can do that to the individual object fields 11 _i , in particular to the individual image fields 23 _i , transferred intensity or dose of the illumination radiation 5 be determined. The sensors 32 _In particular, it is possible to continuously determine the intensity or dose of the illumination radiation 5 , In particular, they make it possible to determine the intensity or dose of the illumination radiation 5 Real time.

The sensors 32 _i are in signal transmitting manner with a control device 33 connected. The control device 33 is in turn in signal-transmitting manner with the coupling-out optics 9 connected.

With the help of the sensors 32 _i and the controller 33 can the decoupling optics 9 controlled, that is controlled, be controlled.

The control device 33 may include a computing unit. The control device 33 may also be in signal transmitting manner with the radiation source 4 be connected. In particular, it can be used to control the radiation source 4 emitted intensity of the illumination radiation 5 serve. For this purpose, a separate control device may be provided.

An FEL is particularly suitable for a projection exposure system 1 with a variety of scanners 3 _i , because the cost of a FEL increases only slowly with its performance, so that the cost per power for a high-performance FEL is significantly lower than for a lower-performance FEL.

According to the invention, it has been recognized that the radiation dose which the individual scanners 3 _i reached, must be very carefully controlled. The allowed error or the allowed fluctuations of the radiation dose, which the individual scanners 3 _i , is in particular at most 1%, in particular at most 0.5%, in particular at most 0.3%, in particular at most 0.2%, in particular at most 0.1%, in particular at most 0.05%.

In particular, the dose must be controlled on a time scale which is faster than the time during which a point on the reticle 22 _i that of the illumination optics 17 _i illuminated area of the object field 11 _i goes through. This time is identical to the time that the optically conjugate point on the wafer 25 _i needed to get the appropriate area of the image box 23 _i go through. This area is also called scan slot. This time is usually at least 1 ms, in particular at least 5 ms. The dose with which a point on the wafer 25 _{i is} illuminated, results from the integral of this movement and is therefore also referred to as scan integral. The dose control takes place in particular on a time scale of at most 1 ms, in particular at most 0.5 ms, in particular at most 0.3 ms, in particular at most 0.2 ms, in particular at most 0.1 ms.

It was further recognized that it is advantageous for such a rapid dose control if the mechanical components used for this purpose need not be repeatedly accelerated and decelerated.

With the means described below for adjusting the radiation dose, which to the individual scanners 3 _i is a variation of the intensity, in particular a variation of the individual scanners 3 _i , in particular their object fields 11 _i , in particular their image fields 23 _i , reaching illumination radiation of up to ± 1%, in particular of up to ± 5%, in particular of up to ± 10% possible.

On longer time scales, a greater dose adjustment may be necessary. This can optionally be achieved with additional resources.

The variants described below exploit the fact that the illumination radiation emitted by a FEL 5 , in particular their intensity, for example by changing the number of electrons generated, by changing the repetition frequency of the FEL or by omitting individual pulses of the FEL very quickly and easily controlled, in particular can be controlled. The control and in particular control frequency of the radiation source 4 is in particular at least 1 MHz, in particular at least 10 MHz, in particular at least 100 MHz. It can be up to 1 GHz, in particular up to 10 GHz, in particular up to 100 GHz. The electrons can be generated in particular by bombarding an electrode with a laser. Lasers can be controlled very quickly and hysteresis-free, so that the intensity is advantageously controllable when using such an electron source.

To the dose for the different scanners 3 _In accordance with the invention, it is possible to set _i independently of one another, the intensity control capability of the FEL and a means for varying the distribution ratio of the total intensity of the collective output beam 8th in single intensities of the single output jets 10 _i used. The total intensity of the FEL and the split ratio of the total intensity of the collection output beam 8th in single intensities of the single output jets 10 _i fixed, so the dose for each scanner 3 _i be regulated independently. It has been recognized that it is necessary for independent dose control for multiple scanners 3 _{i is} sufficient if the split ratio on the multiple scanners 3 _i is describable only by a single degree of freedom. It was also recognized that it is necessary for independent regulation of multiple scanners 3 _i may be sufficient if the distribution ratio to the multiple scanners is only temporally variable, but not controllable.

Different variants of the means for varying the distribution ratio of the total intensity of the collection output beam 8th in single intensities of the single output jets 10 _i will be described below.

The following is with reference to the 4 and 5 a variant of the radiation source module 2 , in particular the coupling-out optics 9 , described.

In the figures, an exemplary coupling-out optics 9 represented, with which the collection output beam 8th in four single output jets 10 ₁ to 10 _{4 is} divided. This serves primarily to clarify the inventive concept. The collection output beam 8th may also be in a different number of single output jets 10 _{i be} split. The number of single output jets 10 _In particular, _i corresponds exactly to the number M of scanners 3 _{i of} the projection exposure system 1 ,

The decoupling optics 9 comprises means for varying the division ratio of the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _i . In the in the 4 and 5 variant shown, this agent is used as a separation plate 34 , which is also commonly referred to as a separation component formed.

The separation plate 34 is rectangular. It is stationary in the beam path of the illumination radiation 5 arranged.

The separation plate 34 is designed in particular as a structured mirror. It has different areas 35 _i up.

Each of the areas 35 _i is exactly one of the scanners 3 _i assigned. The areas 35 _In particular, _i are each designed such that they are responsible for transferring the illuminating radiation impinging on them 5 to a specific one of the scanners 3 _i lead. With the help of the separation plate 34 is thus the collection output beam 8th into the single output jets 10 _i divisible.

The areas 35 _In particular, _i have different orientations. This causes the collection output beam 8th on the different areas 35 _{i has} different angles of incidence. This makes it possible in a simple manner, the collection output beam 8th in the individual scanners 3 _i specifically assigned single output jets 10 _i split up.

In addition, the coupling-out optics 9 a rotating polygon mirror 36 on.

The representation of the polygon mirror 36 in the figures is to be understood by way of example. It is merely an illustration of the underlying concept. The polygon mirror 36 typically has a larger number of side surfaces in reality.

In this variant, the Auskoppeloptik includes 9 thus a rotating element, the polygon mirror 36 , as well as a stationary element, the separation plate 34 , which with a location-dependent distribution ratio of the total intensity of the collection output beam 8th to the individual intensities of the single output jets 10 _{i is} provided.

The rotating polygon mirror 36 is in the beam path between the radiation source 4 and the separation plate 34 , in particular between the beam shaping optics 7 and the separation plate 34 arranged.

The polygon mirror 36 is so in the beam path of the illumination radiation 5 arranged that the collection output beam 8th each from one of the reflective sides of the polygon mirror 36 to the separation plate 34 is reflected where he has an area of illumination 37 illuminates. The illumination area 37 is formed strip-shaped. Alternative forms of illumination 37 are also possible and will be described in more detail below. The exact shape of the illumination area 37 can by means of the beam shaping optics 7 and / or the polygon mirror 36 be determined.

In the in the 4 and 5 illustrated variant, the illumination range extends 37 with its longer side over the entire width of the separation plate 34 , In the direction perpendicular to this covers the illumination area 37 only a portion of the separation plate 34 , The illumination area 37 in particular has a width whose ratio to the length, that is to the longer side, of the separation plate 34 is at most 1: 2, in particular at most 1: 3, in particular at most 1: 5, in particular at most 1:10.

The polygon mirror 36 is rotatably mounted. It is rotatably mounted in particular about a central longitudinal axis, which runs parallel to all its side surfaces.

To the displacement, that is to the rotation, of the polygon mirror 36 is a control unit, not shown in the figures with an actuator, in particular in the form of a motor provided. The rotation of the polygon mirror 36 , in particular the rotational speed thereof, can be controllable. In an alternative embodiment, the polygon mirror 36 a fixed, non-controllable rotational speed. The rotational speed of the polygon mirror 36 does not necessarily have to be kept precisely precise. The polygon mirror 36 can also have a slightly fluctuating rotational speed, provided that there is a means to determine the current position of the polygon mirror.

A rotation of the polygon mirror 36 in the direction of rotation 38 leads to a shift, in particular a shift, the illumination area 37 in a direction of displacement 39 relative to the separation plate 34 , By means of the polygon mirror 36 can thus the illumination area 37 on the separation plate 34 be moved. The relative position of the illumination area 37 on the separation plate 34 and thus the variation of the distribution ratio of the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _In particular, _i can be changed periodically with a period T _Var .

To a variation of the angles of incidence of the collection output beam 8th on the different areas 35 _i with a shift of the illumination area 37 relative to the separation plate 34 To compensate, the areas can 35 _{i be provided} with a refractive power. It can be provided in particular, the areas 35 _i along the in 5 shown x-axis, that is parallel to the direction of displacement 39 to provide a refractive power. Alternatively or additionally, as in 5 is shown schematically and exemplarily, for each of the individual output beams 10 _i an additional optical element 40 , In particular in the form of a mirror, be provided, which for the transfer of each one of the individual output jets 10 _i from the separation plate 34 to one of the scanners 3 _i serves. This optical element 40 can each be provided with a refractive power. The optical elements 40 form components of an imaging optic 54 , The imaging optics 54 , in particular the optical elements 40 can be components of the coupling optics 9 form. With the help of the optical element 40 can be ensured that the entry point of the single output beam 10 _i in the scanner 3 _i independent of the impact point of the collective output beam 8th on the separation plate 34 , in particular independent of the position of the illumination area 37 in the direction of displacement 39 , is.

During a rotation of the polygon mirror 36 the angle of incidence of the collection output beam changes 8th on the reflective side surfaces. To a good approximation, the difference of the largest and the smallest angle of incidence of the collection output beam scales 8th on the polygon mirror 36 with the reciprocal of the number of side surfaces of the same. The more side surfaces of the polygon mirror 36 The lower the interval of the different angles of incidence of the collecting output beam 8th on the polygon mirror 36 , The angle of incidence of the collection output beam 8th on the polygon mirror 36 has an influence on the reflectivity. To make sure the polygon mirror 36 has a minimum reflectivity, a maximum allowable angle of incidence may be specified. Conversely, a minimum angle of incidence can also be specified. With a difference between the maximum angle of incidence and the minimum angle of incidence of 10 °, a number of side surfaces of the polygon mirror resulted 36 from 36.

The number of side faces of the polygon mirror 36 is in particular at least 12, in particular at least 18, in particular at least 24, in particular at least 36. It can be up to 120, in particular up to 180, in particular up to 360.

By increasing the number of side surfaces of the polygon mirror 36 , which are also referred to as facets, the necessary rotational speed of the polygon mirror 36 be reduced.

The larger the number of side surfaces of the polygon mirror 36 , the larger it must be, given the size of the separation plate 34 , the distance between the polygon mirror 36 and the separation plate 34 be. The polygon mirror is in particular designed and arranged such that the illumination area 37 during a rotation of the polygon mirror 36 in each case the entire separation plate 34 sweeps. The angle of rotation of the individual facets thus depends directly on the dimensions of the separation plate 34 in the direction of displacement 39 and the distance between the separation plate 34 and the polygon mirror 36 together. For a polygon mirror 36 with 64 facets and a distance of 1 m between polygon mirror and separation plate 34 results for the direction of relocation 39 swept area a length of about 200 mm. To a good approximation, this information can be scaled for other geometries, that is, the length of the displacement direction 39 swept area is proportional to the distance between polygon mirrors 36 and separation plate 34 and inversely proportional to the number of facets of the polygon mirror.

Between the polygon mirror 36 and the separation component 34 may according to an alternative of the invention, a further component, in particular in the form of a mirror 41 be arranged. The mirror 41 is used to magnify the through the polygon mirror 36 caused deflection angle.

An increase in the deflection angle may be desirable to ensure that the illumination area 37 during a rotation of the polygon mirror 36 the entire separation plate 34 sweeps. As a result, the formation and / or arrangement of the separation plate 34 by the number of facets of the polygon mirror 36 be decoupled.

In addition, by means of the mirror 41 the maximum deflection angle, by which the illumination radiation on its way to the separation plate 34 is deflected in a single reflection can be reduced. By distributing a total deflection to a plurality of partial reflections, the transmission can be increased.

A corresponding alternative in which the mirror 41 in the beam path between the polygon mirror 36 and the separation plate 34 is arranged is schematically in the 43 shown. In the illustrated variant, the collective output beam hits in the illustrated position of the polygon mirror 36 at an angle α to a side surface of the polygon mirror 36 on, so is deflected by a total of 2α.

The mirror 41 is so relative to the polygon mirror 36 arranged that on the side surface of the polygon mirror 36 reflected collection output beam 8th at the mirror 41 is deflected by the angle 2α.

The mirror 41 has a curved surface. Should the mirror 41 just lead to a doubling of the deflection angle, resulting in the shape of the surface of the mirror 41 as solution of the following differential equation:

This differential equation has a one-parameter set of solutions of the following form:

If the deflection angle is to be increased generally by a factor f, the shape of the surface results as a solution of the following differential equation:

The mirror 41 with such a surface forms an angular magnification component with a given translation. In the case of the angular magnification component, the effective incidence and angle of emergence are in particular proportional to one another.

In principle, the mirror 41 also have an alternative surface shape. This is especially possible if the proportionality between the angle of incidence and the angle of failure does not matter.

In principle, the factor f can also be less than 1. In this case, it is an angle reduction component, which is generally regarded as an angular magnification component with a magnification <1.

In the in the 4 and 5 illustrated embodiment performs a rotation of the polygon mirror 36 not only for shifting the illumination area 37 along the direction of displacement 39 , but also to a change in the angle of incidence of the collecting output beam 8th on the separation plate 34 , Such a change in the angle of impact may be undesirable. In other words, it may be advantageous to be able to offset the collective output beam in parallel. The illumination area 37 can then in the direction of displacement 39 relative to the separation plate 34 be shifted without causing a change in the angle of incidence of the collective output beam 8th on the separation plate 34 comes.

A variant of such a solution is schematically in the 41 shown. According to this variant is in the beam path between the polygon mirror 36 and the separation plate 34 an actuatable plane mirror 42 arranged. To clarify the principle, two different beam paths of the illumination radiation 5 with different positions of the polygon mirror 36 and the actuatable plane mirror 42 shown.

An alternative to this is in the 42 shown. According to this alternative, instead of the actuatable plane mirror 42 an optical component with refractive power, in particular in the form of a mirror 43 , intended. The mirror 43 has a radiation - reflecting surface whose shape is in the plane of the drawing 42 can be described by a parabola. In the direction perpendicular to the plane of the drawing 42 can the mirror 43 be formed curvature-free. The radiation-reflecting surface of the mirror 43 may in particular have the shape of a general cylinder jacket portion. As a general cylinder in this case one of two parallel, flat, congruent surfaces, namely a base and a top surface, and a lateral surface limited body referred to. The lateral surface is formed in particular by parallel straight lines. The lateral surface can be generated in particular by shifting a flat curve along a straight line which is not in the plane of the curve. The at the mirror 43 reflected single output jets 10 _i run in particular parallel or at least substantially parallel to a predetermined direction. The mirror 43 is stationary in the beam path of the illumination radiation 5 arranged.

The mirror 43 is relative to the polygon mirror 36 preferably arranged such that the impact point P of the illumination radiation 5 on the polygon mirror 36 is just in the focus of the parabola. The focal length f of the parabola then results from the desired parallel offset amplitude divided by the deflection angle interval of the polygon mirror 36 ,

Since the collection output beam 8th in the direction perpendicular to the propagation direction, but in the plane of the 42 has a finite extent, not all of the mirror 43 reflected rays leave the parabola paraxial. It is therefore advantageous to use the collection output beam 8th form so that it has the smallest possible extent in this direction. It can be directed perpendicular to the drawing plane 42 have a large extent. Such shaping of the collection output beam 8th can in particular by means of the beam shaping optics 7 be achieved. The extent of the collection output beam 8th in the drawing plane of 42 is advantageously less than 1/5 of the focal length of the parabola of the mirror 43 , in particular less than 1/10, in particular less than 1/25 of the focal length. The extent of the collection output beam 8th perpendicular to the plane of the 42 is advantageously at least 5 times the size of the extension in the drawing plane, in particular at least 10 times as large, in particular at least 20 times as large.

The collection output beam 8th is from the beam shaping optics 7 , that is, in particular before impinging on the coupling-out optics 9 , shaped so that it has an elongated cross-section. The cross section of the collection output beam 8th when hitting the decoupling optics 9 In particular, it has an aspect ratio defined by the ratio of the lengths of a long axis and a short axis of the cross section of the collection output beam 8th , of at least 2: 1, in particular at least 3: 1, in particular at least 5: 1, in particular at least 10: 1.

The direction of displacement 39 and the orientation of the long axis of the cross section of the collection output beam 8th are oriented in particular perpendicular to each other.

In the following it is assumed that the intensity of the illumination radiation 5 in the illumination area 37 not from the position in the direction perpendicular to the direction of displacement 39 depends. This can be achieved, for example, in that the collection output beam 8th was homogenized in this direction, that is in the y-direction. This can be done, for example, by means of the beam shaping optics 7 be achieved.

The assumption that the collection output beam 8th in the vertical direction, that is in the direction perpendicular to the direction of displacement 39 , is homogenised, is not necessary for the implementation of the general concept. A vertical edge drop in the intensity of the collection output beam 8th can be compensated, for example, by the areas 35 _i at the top and bottom of the separation plate 34 that is perpendicular to the direction of displacement 39 , be made wider to pick up or reflect the same amount of radiant energy.

The following is based on the 32 . 33 . 34A and 34B described as an example, such as the intensity of the radiation source 4 can be changed depending on the desired doses and on the temporal course of the distribution ratio.

For the sake of simplicity, only the case of dividing an incident beam into two single output beams is shown in the figures 10 _i , 10 _{i + 1} shown. This can be as a section of a division of the collection output beam 8th into a larger number of single output jets 10 _i understood.

In the 32 is an example of a section of the separation plate 34 with two areas 35 _i , 35 _{i + 1} and a lighting area 37 , which in the direction of displacement 39 relative to the separation plate 34 is moved, shown. To clarify the idea, the different areas are 35 _i , 35 _{i + 1} hatched in different dashed lines. The dashing is in the 33 resumed. In this figure is shown how the distribution ratio of the intensity of the incident beam to the two single output beams 10 _i , 10 _{i + 1} changes when the illumination area 37 periodically along the direction of displacement 39 relative to the separation plate 34 is relocated. In the 33 is also exemplified the duration of a period T _{Var of} the variation of the division ratio.

Is the target dose for the scanner 3 _i greater than the target dose for the scanner 3 _{i + 1} , so must the power of the radiation source 4 For example, as exemplified in the 34A represented, controlled. In particular, it assumes a maximum at the times when the ratio of the single intensity of the single output beam 10 _i to the single intensity of the single output beam 10 _{i + 1 is} maximum. Conversely, it assumes a minimum at the times when this ratio is minimal.

Conversely, the target dose for the scanner 3 _i less than the target dose for the scanner 3 _{i + 1} , so must the power of the radiation source 4 For example, as exemplified in the 34B represented, controlled. In this case, the power takes a minimum especially at the times when the ratio of the single intensity of the single output beam 10 _i to the single intensity of the single output beam 10 _{i + 1 is} maximum. Conversely, the power of the radiation source decreases 4 maximum when this ratio is at a minimum.

The in the 34A and 34B exemplified graphs of the power of the radiation source 4 are purely exemplary to understand. By suitable control of the power of the radiation source 4 depending on the distribution ratio of the collective output beam 8th on the single output jets 10 _In principle, as shown, it is possible to use any nominal dose for all scanners 3 _i adjust. This is possible in particular if the number k of the subintervals into which the period T _{Var is} decomposed is sufficiently large. This is possible in particular if the number k of sub-intervals is at least as large as the number M of scanners 3 _i .

Below are different variants of the separation plate 34 based on the exemplary 6 to 16 described.

In the 6 is schematically a separation plate 34 with four areas 35 ₁ to 35 ₄ shown. The different areas 35 ₁ to 35 ₄ serve as previously described the transfer of separate single output jets 10 ₁ to 10 ₄ to 4 different scanners 3 ₁ to 3 ₄ .

The areas 35 _i each have an extension in the y-direction, which depends on its position in the x-direction. Here, the x-direction is opposite to the direction of displacement 39 , The y-direction is perpendicular to the direction of displacement 39 ,

Unless the separation plate 34 has a constant reflectivity over its entire extent, is the single intensity of the different areas 35 _i to different scanners 3 _i reflected single output jets 10 _i directly dependent, in particular directly proportional to the partial area of the area 35 _i , which from the collection output beam 8th is illuminated. This subarea is straight through the intersection, that is, through the overlap area, of the illumination area 37 and the respective area 35 _i given. By a shift of the illumination area 37 in the direction of displacement 39 can this area and therefore the single intensity of the single output rays 10 _{i be} varied. In particular, it is possible to divide the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _i , that is the division ratio of the total intensity into the individual intensities, by shifting the illumination range 37 relative to the separation plate 34 to vary.

Exemplary are in the 6 two different arrangements (A, B) of the illumination area 37 on the separation plate 34 shown. The arrangement A corresponds to just a minimum of the single intensity of the single output beam 10 ₁ . The single output beam 10 ₃ is in position A of the illumination area 37 in about a maximum. In position B the situation is exactly the opposite. In this position has the single output beam 10 ₁ maximum during the single output beam 10 ₃ in the position B has a minimum. The single output jets 10 ₂ and 10 ₄ have in the two positions A, B each have an intermediate value.

In this alternative, the separation plate 34 assign the areas 35 _i apart from the straight edges of the baffle plate 34 each boundary 44 _ij , which can be described by trigonometric functions, in particular by sine functions. Here, the indices i, j respectively give the indices of the adjacent regions 35 _i , 35 _j .

The borders 44 _ij in this alternative can be described in particular by continuous functions. The extent of the areas 35 _In this alternative, _i in the y-direction is greater than a minimum value at any point in the x-direction. The minimum value is about 12% of the total extent of the separation plate 34 in the y direction. Alternative minimum values are also possible. In such a design of the separation plate 34 , in particular the division thereof into the different areas 35 _i ensures that for every arrangement of the illumination area 37 relative to the separation plate 34 a minimum of 12% of the total intensity of the collection output beam 8th to each of the individual scanners 3 _{i is} led. In particular, it is ensured that for each displacement position of the polygon mirror 36 that is, at any time of its relocation, each of the single output jets 10 _{i has} a single intensity which is at least 12% of the total intensity of the collection output beam 8th or at least 48% of the mean single intensity of all single output jets 10 _i .

General is the separation plate 34 preferably designed such that a proportion of at least 50%, in particular at least 70%, in particular at least 90%, in particular 100%, of the single output jets 10 _i during at least 50% of the duration of the period T _{Var of} the displacement of the polygon mirror 36 Have single intensities, which is at least 10% of the mean single intensity of all single output jets 10 _i amounts to.

An alternative feature of a preferred embodiment of the separation plate 34 is described below: For each of the single output jets 10 _i can over the duration of the period T _Var, the time profile of the division share, that is what proportion of the intensity of the collection output beam 8th into the relevant single output beam 10 _i is determined. For each of the single output jets 10 _i the mean, variance and standard deviation of the dividing portion can be determined therefrom. The variance is equal to the square of the standard deviation. Preferably, a proportion of at least 50%, in particular at least 70%, in particular at least 90%, in particular 100%, of the individual output jets 10 _{i is} a standard deviation that is less than half the mean of the corresponding division rate.

In general, an advantageous embodiment of the separation plate can be 34 , in particular their division into the areas 35 _{characterize i} as follows:
Let M be the number of scanners 3 _i . Let the x-coordinate, as before, be parallel to the direction of displacement 39 along which the illumination area 37 relative to the separation plate 34 is displaced, and let the y-coordinate be perpendicular to it, in particular the coordinate along which that of the collection output beam 8th illuminated illumination area 37 is stretched. For the following consideration, the dimensions of the separation plate 34 normalized to 1 in the x and y directions. This can be achieved by a simple scaling.

These assumptions result in the average, ie in the x-direction, average y-extent of the individual regions 35 _i to 1 / M.

Let a be the maximum y-expansion and b the minimum y-extension of a certain range 35 _i , it follows: a ≥ 1 / M ≥ b.

Finally, let A be the proportion of the x-coordinates at which the y-extension of a certain range 35 _{i is} greater than a / 2 + 1 / (2M) and let B be the proportion of the x-coordinates of this range 35 _i , where the y- Expansion is smaller than a / 2 + 1 / (2M). Then advantageous separations of the separation plate can be characterized by the following properties: A + B <½ for M ≤ 8 and A + B <4 / M for M ≥ 8.

Illustrated clearly indicates the y-expansion of the areas 35 _i over the extent in the x-direction as far as possible only a little varying, in particular a substantially constant value and has only in a few places outbreaks up or down.

In 7A is schematically an alternative embodiment of the separation plate 34 shown. In the 7A illustrated embodiment of the separation plate 34 serves to divide the collection output beam 8th in three single output jets 10 ₁ , 10 ₂ and 10 ₃ . The separation plate 34 here has three areas 35 ₁ , 35 ₂ and 35 ₃ on. In this embodiment, the deviation of the partial area of the illumination area 37 covered part of the individual areas 35 _i from a given nominal value depending on the position of the illumination area 37 in the direction of displacement 39 each described by a sum of trigonometric functions. The nominal value may, in particular, be an average, which is a uniform distribution of the collective output beam 8th in single output jets 10 _i corresponds.

The following are further aspects of the coupling optics 9 and the method for controlling the distribution of the illumination radiation 5 on the scanner 3 _i described.

In general, the coupling optics is used 9 in addition, the illumination radiation 5 from the radiation source 4 , in particular the collection output beam 8th , which in particular of the beam-shaping optics 7 from the raw stream 6 can be shaped on the single output jets 10 _i which to the individual scanners 3 _i split up. With the help of the coupling-out optics 9 in particular, the distribution ratio of the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _i varies, in particular controlled, in particular regulated. This allows the distribution of the total energy of the illumination radiation 5 on the individual scanners 3 _i vary, especially control, in particular regulate.

The distribution of the total energy of the illumination radiation 5 on the individual scanners 3 _i can be described by a function f _i (p) which, depending on a parameter p, indicates the proportion of the total energy which results in the ith scanner 3 _{i is} led. The parameter p denotes in a generalized form the setting of the coupling-out optics 9 , in particular the relative arrangement of the illumination area 37 on the separation plate 34 or the setting of the elements of the coupling-out optics 9 dependent transmittance thereof for the different single output jets 10 _i . The change in the distribution ratio of the total intensity to the individual intensity can thus be described by a change in the parameter p and associated changes in the functions f _i (p).

For the parameter p, a time course p (t) is set. Since a change of p is often associated with a mechanical displacement, it is often advantageous and / or even compelling if p (t) is a continuous function. For the inventive realization, however, it is sufficient if the parameter p is set to a certain value for a certain period of time and then instantaneously changed to another value. Since the latter approach to temporally changing p facilitates the description, it will be used below.

The parameter p is set for a certain period of time. In the scan integral, the averaging over the different time periods results. It could be shown that the possibility of temporal variation of the parameter p, in particular by the possibility of choosing the parameter p differently in different time periods, makes the number of degrees of freedom large enough to be the distribution of the total energy averaged over one period all scanners 3 _i essentially free to choose. Practically, this can be done by the illumination area 37 relative to the separation plate 34 is shifted, in particular by successively a sequence of different illumination areas 37 on the separation plate 34 be illuminated. By changing the relative position of the illumination area 37 on the separation plate 34 can the division ratio of the total intensity of the collection output beam 8th to the individual intensities of the single output jets 10 _{i be} changed.

For example, different solutions differ in whether the individual intensities of the single output jets 10 _i can only be attenuated on the basis of an average standard intensity of the same, or whether a redistribution of energy between different scanners 3 _{i is} possible. This can generally by a so-called lever of the coupling-out optics 9 to be discribed. This lever is referred to below by the letter m.

Without actuation of the coupling-out optics 9 becomes through the coupling-out optics 9 in each case a fixed reference component of the power of the collective output beam 8th on a particular one of the single output jets 10 _i steered. For M single output jets 10 _i this proportion can be 1 / M, especially if the scanners 3 _{i are} identical and on them identical processes for exposing a wafer 25 _{i be} carried out. The reference proportion can also be unequal 1 / M, especially if the amount of scanner 3 _i consists of different types. The lever m is the maximum relative deviation of the adjustable portion of a single output beam 10 _i from the reference portion of this single output beam 10 _i denotes. In the following, the amount of the maximum relative deviation is referred to as the lever m, that is, the specification of a certain value of the m does not necessarily include the information that both an increase of the component by m and a reduction of the component by m is possible.

It was further recognized that when redistributed to the individual scanners 3 _i guided illumination radiation 5 an increase to a certain one of the scanners 3 _i necessarily lead to a reduction of the overall amount to the other scanners 3 _j , j ≠ i led amount of energy, provided that the total intensity of the collection output beam 8th is constant.

wobei M die Gesamtzahl der Scanner 3 _i angibt.The task, a predetermined number M of scanners 3 _i with a predetermined scan-integrated dose y of illumination radiation 5 can be mathematically described and solved with the help of the functions f _i (p). Will the scan period, that is, the time that is a point on the wafer 25 _i needed to go through the scan slot, divided into k intervals, and becomes the total intensity of the collection output beam 8th denoted by z _k in the k-th interval, the radiation doses y _i , which become the ith scanner, can be obtained 3 _i , express as follows:

where M is the total number of scanners 3 _i indicates.

Unless the redistribution of energy between each scanner 3 _i the total amount to the scanners 3 _i guided energy of the illumination radiation 5 constant, applies regardless of the distribution of the collective output beam 8th into the single output jets 10 _i , that is independent of the setting of the coupling-out optics 9 , which is described by the parameter p:

According to the invention, it was possible to show how the coupling-out optics 9 has to be designed so that there is a solution {p _k }, ie a sequence of the parameter values p in the k intervals, for a substantially arbitrary predetermined quantity of desired doses {y _i } in order to be able to set these desired doses. Mathematically, this is expressed by the functions {f _i }.

In addition, it was possible to show how the decoupling optics 9 can be controlled to calculate a solution {p _k } for given functions {f _i } from the desired boxes {y _i }.

It could be shown that in order to set M given desired doses, that is to set predefined, scanner-specific desired doses in a projection exposure system 1 with M scanners 3 _{i is} sufficient to divide the scanning period into M intervals k.

A simple variant, such as the distribution of the collective output beam 8th can look for M scanner 3 _i adjust any desired doses, is to divide the Scan Interval into M disjoint intervals and the functions f _i (p _k) to be chosen such that the functions f _i (p _k) p _i be suitably selected in each case in exactly one interval during are constant in the remaining intervals p _j , j ≠ i.

Advantageously, the functions f _i (p) are continuous, at least for the actually used values of p, there can be advantageously achieved so that, in the actual setting of the parameter p, so a shift of the illumination area 37 relative to the separation plate 34 , By means of an actuator small deviations from a target value not to large deviations of the actual distribution ratios of the collective output beam 8th on single output jets 10 _{i lead} from the target split ratio. The sum of all functions f _i (p) is less than or equal to 1 for each value of p, advantageously equal to 1. A corresponding distribution of the total intensity of the collective output beam 8th to the individual intensities of the single output jets 10 _i , that is, a corresponding course of the functions f _i (p) is exemplary in the 39 shown.

While with the course of functions f _i (p) according to 39 in a particularly simple manner intensity increases relative to a mean nominal value f of the single intensity of the single output beams 10 _{i are} adjustable, it may also be advantageous to form the functions f _i (p) in the intervals p _k starting from a nominal value f * with excursions in both directions. This is exemplary in the 40 shown.

wobei δ_{i k} = 1 für i = k und δ_ik = 0 für i ≠ k.If we denote the value of p during the i-th interval minus an average value with x _i , so that x _i directly indicates how the single intensity of the single output beam 10 _i , which during this interval to the ith scanner 3 _i is changed relative to this mean x _i , the functions f _i assume the following simple form:

where δ _{i k} = 1 for i = k and δ _ik = 0 for i ≠ k.

An important question is what value the variables x _{i must} maximally assume in order to set a certain range of target values y _i . For practical implementation, this corresponds to the question of how large the setting range of the coupling-out optics 9 must be to a predetermined range of desired doses of the illumination radiation for the different scanners 3 _{i can} adjust. This is described by the size of the lever m introduced above.

If it is assumed without restriction of generality that any y _i should be adjustable with -1 ≦ y _i ≦ 1, then the necessary adjustment range m of the manipulators is given by the following formula:

As described above, the lever m indicates the amount of the maximum relative deviation of the actual split ratio from a reference split ratio. If the relative deviation can be achieved both upwards and downwards, this is called symmetrical. The above formula describes the symmetrical case. At the other extreme, the relative deviation m can only be achieved either up or down. The necessary adjustment range m of the manipulators is summarized in the following table: Number N of disjoint intervals into which the scan period is divided Asymmetric form of the functions f _i (p) Symmetrical form of the functions f _i (p) N = M - 1 No solution m = 2 (M - 1) ² : M N = M m = 2 (M - 1) m = M - 1 Table: Necessary adjustment ranges of the manipulators for a projection exposure system with M scanners

As described, the normalization of the above equation is such that the maximum possible independent change in the proportion of each of the single output jets 10i , integrated over the scan period, equals 1 is set. The necessary lever m is greater than 1 and accordingly indicates how much more the instantaneous split ratio must be able to be changed more than the default values of the split ratios integrated over the scan period may deviate from a reference value. A minimum value of m is therefore clearly specified by the task.

It was recognized that, although M - 1 intervals may be sufficient, it is usually better to divide the scan period into M intervals. This has the particular advantage that then the necessary adjustment range is significantly smaller. In addition, there is always a solution.

The scan period can also be split into more than M intervals. In particular, it can be divided into an integer multiple of M intervals. A lower limit for the minimum duration of the individual intervals results from the speed with which the distribution of the collective output beam 8th into the single output jets 10 _i can be varied.

Corresponds to the number of disjoint intervals into which the period T _{Var of} the variation of the distribution ratio of the collection output beam 8th into the single output jets 10 _i , even the number M of different scanners 3 _{i of} the projection exposure system 1 Finally, there is no significant difference between symmetric and asymmetric design. Although the setting range m is twice as large in the asymmetric embodiment, only values from 0 to m are needed. In the symmetrical version, m is half as large, but values from -m to + m are needed. The equation given above for the adjustment range m of the manipulators guarantees that a solution exists for y _i in a certain range. This solution is even explicitly constructed during optimization. So it is always possible to use this solution also, if at a later time a certain intensity distribution should be set.

While there is usually only a single solution {y _k } at the edges of the allowed manifold {y _i }, in the case of a division of the period T _Var into at least M disjoint intervals for y _i, an infinite set of conceivable ones usually exists within the allowed range Solutions. When selecting the best solution / solutions, mechanical boundary conditions can be taken into account. For example, the desire can be considered to have to manipulate the manipulators as little as possible.

Concrete designs of the correspondingly designed separation plate 34 are exemplary in the 8th and 9 whose description is hereby incorporated by reference. The variant of the separation plate 34 according to 8th has a setting range m = 0.9. The in the 9 variant shown has a setting range m = 2.

According to alternative variants, the functions f _i (p) have a non-linear course. For M = 3, for example, the following functions can be selected as functions f _i (p): f ₁ (p) = cosp, f ₂ (p) = sinp and f ₃ (p) = -cosp - sinp.

An alternative option is the following: f ₁ (p) = cosp, f ₂ (p) = cos (p + 2π: 3) and f ₃ (p) = cos (p + 4π: 3).

For both alternatives:

beziehungsweise für das zweite BeispielIn both alternatives, a division of the period T _Var into two disjoint intervals can be set to a desired energy distribution y ₁ and y ₂ with y ₃ = -y ₁ -y ₂ . Except for a possible permutation of p ₁ and p ₂ , there is a clear solution for the first example

or for the second example

However, there is only one solution if y ₁ , y ₂ and y ₃ = -y ₁ -y _{2 are} within certain limits.

It could be shown that any y _{i can be set} with -1 ≦ y _i ≦ 1, provided that the manipulator has a lever m, which serves as a pre-factor before the functions f _i (P). The required lever m is summarized for the different cases in the following table:

It could further be shown that the functions f _i (x) = cos ^q (x + c _i ) with c _i = 2πi / (q + 2), i = 1 ... M and q are odd for M = q + 2 scanners 3 _i always represent a solution. In other words, this family of functions {f _i } provides a way of dividing the separation plate 34 for a projection exposure system 1 with M = q + 2 scanner 3 _i on. The form of the functions f _i (p) = cos ^q (p + c _i ) is given in 44A to 44D exemplified for q = 1, 3, 5 and 7. Corresponding divisions of the separation plate 34 in separate areas 35 _i are in the 7A to 7D shown.

It becomes clear that the larger the exponent q, the more the functions f _i (p) are located. Practically, this means that the individual scanners 3 _The more isolated the functions f _i (p) are, the better they can be controlled independently of each other.

In the 8th and 9 are other alternatives of the separation plate 34 shown. These alternatives are the boundaries 44 _{i j} each formed linearly in sections. The separation plate 34 points in the direction of displacement 39 essentially four areas each, in which the relative proportions of the areas 35 _i vary constantly. Between in the direction of displacement 39 adjacent areas There is a sudden change in the extent of the areas 35 _i in the direction perpendicular to the direction of displacement 39 ,

The alternatives according to 8th and according to 9 differ by the value of the adjustment range or lever m. The adjustment range m thus forms a measure of how much the different divisions of the total intensity of the collection output beam 8th in single intensities of the single output jets 10 _i can deviate at most from a uniform distribution. It results directly from the training of the coupling optics 9 , in particular from the formation of the separation plate 34 , in particular their division into the different areas 35 _i . Because the coupling optics 9 , in particular the interaction of their relatively displaceable components, for varying or generally manipulating the distribution ratio of the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _i , the setting range m also becomes the setting range m of the coupling-out optics 9 or designated as the adjustment range m of the manipulators. At the in 8th The alternative shown is the adjustment range m = 0.9. At the in 9 shown alternative is m = 2.

In a particularly advantageous embodiment, which is shown schematically in FIG 10 is shown, are the individual areas 35 _i in the direction perpendicular to the direction of displacement 39 into a plurality of subareas 35 _ij split.

In the 11 to 16 are schematic three-dimensional grid drawings of different alternatives of the separation plate 34 exemplified.

In the in the 11 and 12 illustrated alternatives, the separation plate 34 Kinks, but no jumps.

In the in the 13 to 16 alternatives shown have the separation plates 34 Jumps on.

At the in 11 . 13 . 15 illustrated alternatives is the illumination area 37 on the separation plate 34 oriented horizontally, that is, the entire width of the separation plate is illuminated, but only part of the height. At the in 12 . 14 and 16 illustrated alternatives is the illumination area 37 on the separation plate 34 oriented vertically, that is, the entire height of the separation plate is illuminated, but only part of the width.

At the in 11 . 13 . 14 The alternatives shown differ the direction of the single output jets 10 _i to various scanners by the orientation in a horizontal plane, that is, the single output jets 10 _i to different scanners 3 _i run side by side. At the in 12 . 15 . 16 The alternatives shown differ the direction of the single output jets 10 _i to different scanners 3 _i by the orientation in a horizontal plane, that is, the single output beams to different scanners run one above the other.

Other configurations are possible, in which, for example, the orientation of the illumination area 37 on the separation plate 34 are not parallel to one of the axes. In other embodiments, the plane in which the directions of the single output beams lie 10 _i to different scanners 3 _i are not parallel to a coordinate plane. In other embodiments, the directions of the single output jets are 10 _i to different scanners 3 _i not in a plane.

The following are general aspects of the method according to the invention for controlling the distribution of the illumination radiation 5 on the scanner 3 _i with the aid of the coupling-out optics 9 described. Although the description of these aspects is based on individual concrete embodiments of the coupling-out optics 9 The considerations apply mutatis mutandis to all of the embodiments shown in this invention.

It is used in the following without limitation of generality the convention that at each of the scanners 3 _{i is} a dose of the illumination radiation 5 of 1 per unit of time would arrive, provided the illuminating radiation 5 from the radiation source 4 without additional components evenly on the scanner 3 _i would be distributed.

The dose of illumination radiation 5 with which a point on one of the wafers 25 _i is given by the temporal integral of the instantaneous dose of radiation with which this point is exposed as it passes through the scan slot. Within this time, the intensity of the radiation source 4 , in particular the individual intensity of the corresponding scanner 3 _i guided single output beam 10 _{i be} varied.

gegeben, wobei a_{i k} ≥ 0.The radiation source 4 is operated in particular pulsed. It emits illumination radiation 5 in the form of a sequence of pulses with illumination radiation 5 , It is further assumed that the time over which the division ratio of the total intensity is varied to the individual intensities, in particular the period T _{Var of} this variation, is distributed in N-disjoint intervals. The intervals may in particular have identical widths, that is to say identical interval durations. They can also have different widths. Is the sum of the intensities of the radiation source 4 In the kth interval emitted pulses with illumination radiation 5 labeled x _k , so is the total intensity that is the ith scanner 3 _i reached through

given that a _{i k} ≥ 0.

It is assumed that a _{i k} is constant over the length of the k-th interval and / or that the individual pulses each have an identical intensity in the interval k. However, the considerations can easily be extended to the case that the intensity of the coil is variable within the individual intervals. For pulses of constant intensity, the overall intensity of the illumination radiation results 5 which is the ith scanner 3 _i in the k-th interval, from the intensity of a single pulse, in particular the mean intensity of a single pulse, which is weighted by the length of the interval.

Depending on whether the individual areas 35 _i the separation plate 34 only a weakening effect with a direction of displacement 39 have varying attenuation, or whether the areas 35 _i to a redistribution of illumination radiation 5 between the individual scanners 3 _i lead, different limits for the matrix elements a _{i k} can be specified.

The elements a _{i k of} the matrix A essentially indicate which radiation dose of the illumination radiation 5 in the time interval k the scanner 3 _i reached, provided the radiation source 4 in this interval illumination radiation 5 emitted with the power 1.

Die einzelnen Matrixelemente a_{i k} können größer als 1 sein. Es kann jedoch eine Obergrenze u angegeben werden: 0 ≤ a_{i k} ≤ u.Will that of the radiation source 4 maximum emitted total intensity without restriction of generality is standardized to 1, generally applies:

The individual matrix elements a _{i k} can be greater than 1. However, an upper limit u can be given: 0 ≦ a _{i k} ≦ u.

Generally, it is not necessary that there is only a single separation plate 34 indicates where the radiation is to multiple scanners 3 _{i is} split. Another possibility is to first radiate the radiation according to a fixed ratio on single output beams 10 _i split, and the single output jets 10 _i then each lead over its own separation plate.

In particular, in each case only a variable part of the single output beam can be attached to the separation plates 10 _i to the scanner 3 _i while removing some of the energy from the single output beam. This alternative embodiment is based on the 28 and 29 described in more detail below.

In the embodiment with a single separation plate 34 was the split ratio f _i (p) on the single output jets 10 _i to the scanners 3 _{i described} by a parameter p. Are all scanners 3 _i equals, so the f _i (p) can be chosen so that - if the two edges of the separation plate 34 are identified with each other, that is, at the top and bottom of the separation plate, the division ratio to the single output jets 10 _i is equal to -f _{i + 1} (p) = f _i (p + Δ). In other words, the different functions f _i (p) differ only by a shift of the parameter p. If the parameter p is changed periodically, then the different functions f _i (p) differ by a phase shift. 6 is an example of such a design.

Furthermore, it is also assumed in an embodiment with a plurality of separation plates that these separation plates are constructed identically and are phase-shifted relative to one another only in their control.

In addition, it is further assumed that the period T _Var at M scanners 3 _{i is divided} into N = M-disjoint intervals, wherein the radiation source in each of the M intervals a certain, identical number of pulses with illumination radiation 5 emitted.

The matrix A is a cyclic matrix under the above assumptions. This is advantageous because linear arithmetic operations, such as solving systems of equations, for such matrices can be performed very efficiently by returning to one-dimensional discrete Fourier transforms.

The matrix A is in particular invertible, that is, there exists a matrix B with matrix elements b _{i k} , so that:

In the following, y _max denotes the maximum intensity of all the single output jets 10 _i , that is the maximum intensity that a scanner 3 _{i should} receive. Accordingly, y _min denotes the minimum intensity of all the single output jets 10 _i . The ability to set different intensities for each of the scanners 3 _i is determined by the ratio of y _max and y _min . In the following, to characterize the range of the different divisions of the total intensity of the collection output beam 8th in single output jets 10 _{i is} characterized by the parameter r = (y _max -y _min ) / y _max .

By means of a numerical optimization, the dependence of the parameter r on y _max for different numbers M of the scanners could be determined 3 _{i be} examined. Further analysis of the results revealed the following general context:

This makes it possible to determine the maximum value y _max for which a certain value of r is adjustable. Without a weakening in the beam path, the total intensity y _{i would be} at the scanner 3 _i proportional to the total number N of intervals, so that the ratio y _max / N indicates the energy efficiency. By inverting the equation for r _N (y _max ) given above, the energy efficiency can be determined. The result is in the 17 for different values of r as a function of the number M of scanners 3 _i shown.

Generally, by means of the coupling-out optics 9 , in particular by means of the separation plate 34 , Energy of the illumination radiation 5 between the scanners 3 _{i be} redistributed. By this is meant in particular that starting from a mean energy or individual intensity of the individual output beams 10 _{i is} the single intensity of a subset of the single output jets 10 _i at the expense of the individual intensities of another subset of the single output jets 10 _i can be increased. In extreme cases, the entire collection output beam 8th that is, the total intensity of the collection output beam 8th , at a given time in each case to exactly one of the scanners 3 _{i be} led. The collection output beam 8th in other words, in each case on exactly one of the single output jets 10 _i , while the rest of the single output jets 10 _j , j ≠ i have single intensities = 0 at this time. In this case, the dose of the scanner 3 _i by controlling, in particular regulation, the power of the radiation source 4 controlled, in particular, regulated, without affecting the other scanner 3 _j has. However, in this case, each of the scanners 3 _i only to a portion 1 / N of the period T _Var with illumination radiation 5 provided. This can lead to pulse quantization effects.

The parameter u indicates in particular the maximum value of the matrix elements a _{i k} . Clearly, this value gives the ratio of the maximum single intensity of one of the single output jets 10 _i to the average single intensity of all single output jets 10 _i , which is just the total intensity of the collection output beam 8th divided by the number M of scanners 3 _i equals, again. It is assumed that the total intensity of the collection output beam 8th completely on the single output jets 10 _{i is} distributed.

In the 18 to 23 is the energy loss (N - y _max ) / N as a function of a parameter u for different numbers M of the scanners 3 _i shown. In the figures, curves for r = 5%, r = 10% and r = 15% are shown by way of example. Like that 18 to 23 is qualitatively removable, an increase in the parameter u leads to a reduction in energy loss. According to the invention is provided in particular, the separation plate 34 , in particular their areas 35 _i , in such a way that u ≥ 2. In particular, u ≦ 6 can be chosen as the upper limit for u.

It can be shown analytically that for a large number N of intervals, the value for achievable energy efficiency 1 / (1 + √ r ² approaches. The use of a gray filter, whose influence on the individual intensities of the single output jets, is therefore an object of the invention 10 _i only by energy absorption, only for small values of M and / or small values of r. The use of a gray filter is envisaged, in particular at a maximum of M = 5, in particular a maximum of M = 3 scanners 3 _i . The use of a gray filter is envisaged in particular at r <0.1, in particular r <0.05, in particular r <0.02, in particular r <0.01.

The following is with reference to the 24 to 26 an alternative to the coupling-out optics 9 described. Identical parts are given the same reference numerals as in the variant described above, the description of which is hereby incorporated by reference.

In this embodiment, instead of the stationary separation plate 34 a rotatable separation plate 45 intended. On the rotating polygon mirror 36 can be omitted in this variant.

In this variant, the Auskoppeloptik includes 9 thus a rotating element, the separation plate 45 , which with a location-dependent distribution ratio of the total intensity of the collection output beam 8th to the individual intensities of the single output jets 10 _{i is} provided.

The separation plate 45 is anular, that is annular, formed.

The illumination area 37 that is, the area which is from the collection output beam 8th is illuminated, is stationary in this variant. By a rotation of the separation plate 45 thus becomes the separation plate 45 relative to the illumination area 37 and thus the area of illumination 37 relative to the separation plate 45 relocated.

The different areas 35 _i have borders each 44 _ij , which are designed such that the illumination of the area 37 covered parts of the areas 35 _i with a rotation of the separation plate 45 change. As a result, the division ratio of the total intensity of the collection output beam 8th in single intensities of the single output jets 10 _{i be} varied. The distribution ratio can be varied in particular periodically. In this case, the period in particular just a whole revolution of the separation plate 45 correspond.

In the embodiment shown by way of example in the figures, the separation plate 45 four different areas 35 ₁ to 35 ₄ on. The number N of areas 35 _i again corresponds to the number M of scanners 3 _i . It is also possible with this variant, the areas 35 _{i divided} into several, separate, that is, non-contiguous sub-areas. The areas 35 _In particular, _i can be subdivided into subareas in such a way that at least one of them belongs to a specific scanner 3 _i guided single output jets 10 _i of several, separate, that is not contiguous subregions on the separation plate 45 is reflected.

In reality, the projection exposure system includes 1 usually more than four scaners 3 ₁ to 3 ₄ . In this case, the number n of the different areas is also 35 _i the separation plate 45 correspondingly larger. In the 26 is schematically illustrated how a division of the collection output beam 8th into the different single output jets 10 _{i is} reached. The individual areas 35 _i have different inclination angles α _i against one to the back of the separation plate 45 parallel plane 46 on. This leads to different directions of failure 47 _i at the individual areas 35 _i reflected, that is deflected, single output jets 10 _i .

To a shadowing of the single output jets 10 _i through neighboring areas 35 _i can avoid the areas 35 _i , as exemplified in the 27 is shown, in the direction perpendicular to the plane 46 be offset. The areas 35 _i are preferably offset in particular from one another such that the separation plate 45 Overall, a continuous surface, that is, a stepless surface having. A appropriate training of the separation plate 45 with a continuous, stepless surface is particularly advantageous if the reflection of the illumination radiation 5 to take place in grazing incidence. Under grazing incidence is here in particular an incidence of the illumination radiation 5 with angles of incidence of at least 70 ° to the surface normal of the reflecting surface understood.

The following is with reference to the 28 another variant of the coupling optics 9 described. Identical parts are given the same reference numerals as in the previously described variants, to which reference is hereby made.

In this variant, the Auskoppeloptik includes 9 a split unit 48 for splitting the collection output beam 8th in single output jets 10 _i . The allocation unit 48 includes a plurality of deflecting elements 49 _i . The deflecting elements 49 _In each case, _i serve to redirect and decouple one of the individual output jets 10 _i from the collection output beam 8th , They are stationary in the beam path of the illumination radiation 5 arranged.

For variation of the individual intensities of the single output jets 10 _In each case elements for influencing the transmission are provided. It is in particular each of the single output jets 10 _i assigned such an element. The element for influencing the transmission is in particular designed in each case as a rotating element with location-dependent transmission. In particular, these may be rotating gray filters 50 _i act. The gray filters 50 _In particular, each have an anular filter region whose absorption and thus transmission rate in the azimuthal direction, that is, as a function of the rotational position, varies.

In the 28 is exemplarily and schematically shown a variant in which each of the individual output beams 10 _{i is} a separate gray filter 50 _{i is} assigned. In principle, it is also possible to have several, in particular all, of the individual output jets 10 _i with a single rotating gray filter 50 to influence. Here, the gray filter 50 each a separate filter area for each of the single output jets 10 _i have. It is also possible the single output jets 10 _i by different impact areas on the gray filter 50 _i different influence, especially mitigate. In this case, it may be sufficient to use the gray filter 50 with a single anular filter area. Other alternatives are also possible. A gray filter can also be realized in reflection by directing a variable portion of the radiation out of the beam path and then absorbing it.

The gray filters 50 _i allow a periodic adjustment of the individual intensities of the single output jets 10 _i . The single intensities of the single output jets 10 _i can in particular with the help of the gray filter 50 _i be controlled independently.

The total intensity of the collection output beam 8th can in turn be controlled as needed, in particular regulated. In particular, depending on the variation of the distribution ratio, it can estimate the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _i controlled, in particular regulated. In particular, it can also be dependent on predetermined desired doses of the illumination radiation 5 which the different wafers 25 _i should be controlled, in particular regulated. For further details, additions and alternatives refer to the previous alternatives.

In this embodiment, the number of rotating elements may be just as large as the number of single output beams 10 _i , especially just as large as the number of scanners 3 _i . The rotating elements are provided with a location-dependent absorption or reflectivity or transmission.

The following is with reference to the 29 another variant of the coupling optics 9 described. Identical parts are given the same reference numerals as in the previously described variants, to which reference is hereby made.

In this variant, the Auskoppeloptik includes 9 instead of the rotating gray filter 50 _i the in 28 variant shown in each case a rotating polygon mirror 51 _i and each one associated with this stationary component with a location-dependent transmission, in particular reflection or absorption. The stationary component is each a rectangular plate 52 _i with in the direction of displacement 39 formed with varying transmittance.

The polygon mirror 51 _i leads in each case to a direction of displacement 39 movable illumination area 37 on the plate 52 _i . The displacement of the illumination area 37 is essentially the same as in 4 illustrated variant. However, the plate serves 52 _i in the variant according to 29 not for splitting the collection output beam 8th in single output jets 10 _i , but only for influencing the single intensity of one of the single output jets 10 _i , which is the scanner 3 _i reached.

In the following, different methods for controlling the distribution of the total intensity of the collection output beam will be described 8th to the individual intensities of the single output jets 10 _i described.

In all of the variants described above, the coupling-out optical system comprises 9 at least one rotatable, that is at least one periodically displaceable element. To control the distribution of illumination radiation 5 the radiation source 4 to the majority M of scanners 3 _i can first of all the temporal course of the distribution ratio of the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _{i are determined} as a function of the periodically changing setting of this element.

Also provided is the mean actual intensities of each of the single output jets 10 _i to determine. The actual intensities can in particular be determined averaged over a period in each case.

It is possible, the actual intensities each at the entrance of the scanner 3 _i and / or in the area of the object field 11 _i and / or in the area of the image field 23 _i to determine. In principle, the actual intensities can also be at other locations in the beam path behind the coupling-out optics 9 be determined.

For given target intensities of the single output beams 10 On the basis of the determined actual intensities, _i can be correction factors for adjusting the average actual intensities of each of the individual output beams 10 _{i be} determined.

Finally, the overall intensity of the collection output beam 8th depending on the determined correction factors and in dependence, in particular in time dependence on the periodic change of the means for varying the distribution ratio of the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _{i be} controlled. This can be done in particular with the help of the control device 33 happen. The total intensity of the collection output beam 8th can in particular by controlling the radiation source 4 , in particular by controlling the pulse frequency of the radiation source 4 , in particular by controlling the average pulse frequency of the radiation source 4 , be achieved. In this case, the control can in particular be fed back. In other words, it can be a regulation.

It can be provided in particular to individually control the pulse frequency of the radiation source for each of the disjoint intervals of the period T _Var , in particular to regulate it. The pulse rate can be constant within the individual intervals. It can also be controlled, in particular regulated, within these intervals.

Instead of disjoint and discrete intervals, the intensity and / or pulse frequency of the radiation source can also be continuously controlled, in particular regulated. This can be particularly advantageous if the distribution ratio of the collection output beam 8th in single output jets 10 _{i is} continuously, especially constantly changing.

Alternatively, the control of the distribution of the illumination radiation 5 from the radiation source 4 on the scanner 3 _i happen as follows: First, target intensities of the single output jets 10 _i or desired cans, with which the wafers 25 _i in different scanners 3 _i are to be exposed.

It becomes a time course of the distribution ratio of the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _{i is determined} as a function of the variable setting of the means for varying this division ratio.

Then, a time sequence p _k of settings of the distribution ratio varying means becomes the total intensity of the collection output beam 8th into the individual intensities of the single output jets 10 _i determines with which one over the entire sequence p _k averaged distribution ratio of Total intensity of the collection output beam 8th on the single output jets 10 _i can be achieved. The temporal sequence p _k is determined in particular as a function of the predetermined desired intensities.

Then, the actual intensity averaged over the sequence p _k becomes each of the single output beams 10 _{i determined} by going through the entire sequence p _{k of} the setting of the means for varying the division ratio.

From this, deviations of the determined actual intensities from the predetermined desired intensities are determined.

An adjustment of the overall intensity of the collection output beam 8th or their time course and / or the sequence of the settings p _k can be done continuously. It can also take place only if at least one of the deviations exceeds a predetermined maximum value.

The total intensity of the collection output beam 8th In particular, it can be time-controlled, in particular regulated. In particular, it can be controlled, in particular regulated, as a function of the sequence p _{k of} the settings of the means for varying the division ratio.

To control the overall intensity of the collection output beam 8th and for controlling the sequence of the settings p _k , two separate control loops are provided in particular. The control loop for controlling the total intensity of the collection output beam 8th , in particular the intensity of the radiation source 4 emitted illumination radiation 5 will be in time, which is a point on the wafer 25 _In order to pass through the scanning slot, it takes at least once, in particular several times, in particular at least three times, in particular at least five times, in particular at least ten times, to pass through. The duration within which a loop of the control loop takes place is designated analogously to the embodiments described so far as T _Var .

The control of the setting of the means for varying the distribution ratio of the collective output beam 8th in single output jets 10 _i can take place in a separate control loop. This is advantageous because at any time from the measurements of the actual intensities of the single output beams 10 _i can come out new default values for the distribution ratios to be set, but at any time only exactly one value for the instantaneous distribution ratio of the radiation source 4 emitted illumination radiation 5 that is the collective output beam 8th , on the single output jets 10 _i can be set. From the default values for the target dose values, the sequence p _{k of} the settings of the distribution ratio distribution means of the collection output beam becomes 8th into the single output jets 10 _i determined. Then, the means for varying the division ratio is set according to the detected sequence p _k .

In the following, further details of these methods as well as further variants of the coupling-out optics are described 9 described in more detail.

The following is with reference to the 30 a further embodiment of the coupling-out optics 9 described. The in the 30 variant shown corresponds essentially to the in the 5 , to the description of which reference is hereby made. Instead of the rotating polygon mirror 36 is in the variant according to 30 a displaceable, in particular an actuatable displaceable deflection mirror 53 intended. By a shift, in particular a tilt, the deflection mirror 53 is the area of illumination 37 relative to the separation plate 34 displaced. By controlling the displacement position of the deflection mirror 53 can thus the division of the collection output beam 8th into the single output jets 10 _{i be} varied.

At a distance between the deflection mirror 53 and the separation plate 34 of 2 m and an extension of the separation plate 34 in the direction parallel to the direction of displacement 39 of 100 mm, the tilting range of the deflecting mirror 53 relative to a middle position ± 12.5 mrad. Such a tilt corresponds to a base of 10 mm, a movement of two linear actuators of ± 62.5 microns around their zero positions. For the displacement of the deflecting mirror 53 Piezo actuators can be used. These allow in particular a very fast displacement of the deflection mirror 53 , The time required for the displacement of the deflection mirror 53 is required between two displacement positions is in particular less than 1 ms, in particular at most 0.1 ms, in particular at most 0.05 ms, in particular at most 0.03 ms, in particular at most 0.02 ms, in particular at most 0.01 ms.

Like in the 31 is shown schematically, the coupling-out optics 9 according to a variant instead of a single deflecting mirror 53 also a plurality of deflecting mirrors 53 For example, two, three, four, five or more deflecting mirrors 53 include. When using a number q of deflecting mirrors 53 can simultaneously q different illumination areas 37 on the separation plate 34 from a single collection output beam 8th be lit up.

An advantage of this alternative is that the actuators for the displacement of the deflection mirror 53 must be moved slower than the alternative with a single deflection mirror 53 , The time between two switching operations can in particular by a factor which just the number q of the deflection mirror 53 corresponds, be longer than de alternative with a single deflection mirror 53 , In addition, the displacement can be less per switching operation. In principle, it is also possible for each of the individual deflection mirrors 53 a separate separation plate 34 assigned.

The number of deflection mirrors 53 in particular, just the number of scanners 3 _i correspond. It is also possible that the coupling-out optics 9 a number of deflecting mirrors 53 which is less than the number of scanners 3 _i .

An embodiment in which the number q of the deflection mirror 53 at least the number M of scanners 3 _i , q ≥ M, offers the advantage that within a period of the control loop, no change in the setting of the deflection mirror 53 must be done. The time scale on which the deflection mirrors 53 must be set, so is no longer given by the time T _Var , but only by the time scale on which the relevant drift effects of the lithographic system 1 take place.

Also with the variants according to 30 and 31 Advantageously, an angular magnification component according to 43 be used. As a result, the range of tiltability of the deflection mirror 53 be reduced.

The shape of the illumination area results from the beam properties of the radiation source 4 as well as the configuration of the beam shaping unit 7 , Like in the 35 is shown schematically, the illumination area 37 not necessarily be strip-shaped, in particular not necessarily rectangular. In the 35 are exemplary two different illumination areas 37 with a round, in particular an elliptical shape shown. The two in the 35 illustrated illumination areas 37 differ in their extent in the direction of displacement 39 ,

In a first approximation, a free electron laser emits a Gaussian radiation profile. However, necessary for the following considerations is merely the more general property that the illumination on the separation plate 34 factorized: I (x, y) = I _x (x) · I _y (y), and that there is a mirror symmetry: I _x (-x) = I _x (x). These conditions are particularly true when both the beam properties of the radiation source 4 as well as the beam shaping properties of the beam shaping unit 7 meet these conditions. For a straight, linear course of the boundary 44 _{i, i + 1} causes this to cause the expansion of the illumination area 37 in the direction of displacement 39 for the division ratio of the collection output beam 8th on the single output jets 10 _i becomes irrelevant. This no longer applies if the course of the boundary 44 _{i, i + 1} , as exemplified in the 36 is shown having a kink and / or a curvature. In this case, the division ratio is not only dependent on the position of the center of the illumination area 37 , but also of its extent in the direction of displacement 39 ,

Advantageously, therefore, in the design of the separation plate 34 tried, kinks of the borders 44 _ij avoid as _much as possible. In addition, bends are preferably kept as small as possible. Radii of curvature of the borders 44 _ij are advantageously greater than twice the diameter of the illumination area 37 , in particular greater than five times the diameter of the illumination area 37 , in particular greater than ten times the diameter of the illumination area 37 ,

In addition, in the control of the division ratio, especially in the control of the displacement of the illumination area 37 , preferably avoided that the illumination area 37 to lie in the range of kinks or areas with large curvatures.

In the 37 is an example of a separation plate 34 with three areas 35 ₁ , 35 ₂ , 35 ₃ shown. The borders 44 _1,2 , 44 _2,3 are piecewise linear in this variant.

In the 37 are the positions in the direction of displacement 39 by circular symbols 55 highlighted to which the area 35 ₃ in the direction perpendicular to the direction of displacement 39 just one third of the total extension of the separation plate 34 having. This corresponds exactly to the positions at which for the scanner 3 ₃ as single intensity of the single output beam 10 _{3 is} the value of the standard intensity, that is, the value that equals to all three scanners 3 ₁ , 3 ₂ , 3 ₃ divided overall intensity. Three of these positions are beneficial because of the boundary 44 _{2.3 has} no kink at these points. These positions are indicated by filled in symbols. The areas therefore advantageously used in this variant for the displacement of the illumination area 37 are by double arrow symbols 56 displayed.

To make these areas as large as possible, it may be provided to split up the non-advantageous areas. This is exemplary in the 38 shown. By changing the edge steepnesses, the advantageously used areas were able to shift the illumination area 37 in the direction of displacement 39 be increased considerably.

The following are further aspects for setting the coupling-out optics 9 portrayed.

In the preceding considerations, it was at least partially assumed that the external parameter p is changed instantaneously at discrete points in time. This made it possible to derive general yet quantitative results. In reality, the impact position of the illumination radiation 5 on the separation plate 34 , in particular the relative displacement position of the illumination area 37 to the separation plate 34 , generally not instantaneous, but constantly changing with time.

If the finite speed of the change in the parameter p is taken into account, significantly more different positions of the illumination range can ultimately be achieved 37 be approached within the period T _Var , but these positions are no longer independently selectable. This can lead to a larger range of desired intensities y _{i being able to be} set with a given lever m.

If the fact is ignored that the external parameter p can only be changed continuously over time, the finite speed results in the parameter p remaining at a target position for a certain time, and then subsequently approaching the next target position at finite speed.

All of the described embodiments fulfill the property that for each area, which increases the single intensity of a particular one of the single output beams 10 _i , there is another area in which the single intensity of this single output beam 10 _{i is} reduced. If parameter p is thus moved once over the entire travel range at a constant speed, the net effect is averaged out.

The finite travel time has, in particular, the effect that the time at which the external parameter p has one of the desired values becomes smaller. This must be taken into account when determining the value of the functions f _i (p) to be set.

It was further recognized that the setting of the external parameter p may be quite inaccurate, since the value of this parameter p is determined in a control loop. Relevant to this is how many passes of the control loop within the time in which a point on the wafer 25 _i passes through the scan slot. As a first approximation, the error of the division ratio averaged over the scan slot is given by the errors of the instantaneous division ratio divided by the root of the evenly described number of passes. The resulting relative error is advantageously less than ½ of the allowable relative error of the dose by which the wafer 25 _{i is} exposed, in particular less than 1/10 of the allowed relative error of the dose.

The in the individual scanners 3 _i entering energy, in particular the individual intensity of the different single-output beams 10 _i , can be measured very accurately and very quickly by means of sensors, in particular by means of energy sensors or dose sensors. Based on these measurements, it can be calculated which dose correction is necessary. The total intensity of the radiation source 4 emitted illumination radiation 5 , in particular the total intensity of the collection output beam 8th , in particular as a function of the target dose values y _i , by means of the control device 33 to be controlled. It can in particular with the help of the sensors 32 be managed.

• – Vorgabe einer Wunschdosis s ^_i für den i-ten Scanner 3 _i,
• – Bestimmten der aktuellen Gesamtintensität l ~ des Sammel-Ausgabestrahls 8. Die aktuelle Gesamtintensität l ~ des Sammel-Ausgabestrahls 8 kann aus dem vorherigen Durchlauf der Regelschleife bekannt sein. Sie kann auch jeweils gemessen werden.
• – Ermittlung der aktuell eingestellten, über die Scandauer gemittelten, relativen Dosisveränderung y ~_i . Dieser Wert kann von den vorherigen Durchläufen der Regelschleife übernommen werden.
• – Messen der tatsächlichen Dosis s_i des i-ten Scanners 3 _i, vorzugsweise als gleitendes Mittel über die Zeit;
• – Abschätzung, welcher Anteil der Gesamtintensität des Sammel-Ausgabestrahls 8 das Retikel 22 _i eines bestimmten Scanners 3 _i erreicht als: s_i/(l ~_i(1 + y ~_i)) .
• – Lösung des folgenden Gleichungssystems
  

To control the overall intensity of the collection output beam 8th the following control loop can be run through:

• - Specification of a desired dose s ^ _i for the ith scanner 3 _i ,
• - Determine the current total intensity of the collection output beam 8th , The current total intensity l of the collection output beam 8th may be known from the previous pass of the control loop. It can also be measured in each case.
• - Determination of the currently set, relative to the Scandauer averaged, relative dose change y ~ _i , This value can be taken from the previous runs of the control loop.
• Measuring the actual dose s _{i of} the ith scanner 3 _i , preferably as a moving average over time;
• - Estimate what proportion of the total intensity of the collection output beam 8th the reticle 22 _i a particular scanner 3 _i reached as: s _i / (l ~ _i (1 + y ~ _i )) ,
• - Solution of the following equation system
  

• – Anpassung der Gesamtintensität des Sammel-Ausgabestrahls 8, insbesondere durch Anpassung der Leistung der von der Strahlungsquelle 4 emittierten Beleuchtungsstrahlung 5 in Abhängigkeit von der ermittelten Lösung für 1.

This equation system always has a solution.

• - Adjustment of the total intensity of the collection output beam 8th in particular by adjusting the power of the radiation source 4 emitted illumination radiation 5 depending on the determined solution for 1.

This control loop must be in time, which is one point on the wafer 25 _i need to go through the scanning slot will run at least once.

It may also undergo several times, in particular at least twice, in particular at least three times, in particular at least five times, in particular at least ten times.

The determined values for y _i can be passed to a second control loop to determine new default values of the external parameter p _k . From the determined values for y _i , new default values of the external parameter p _k can also be determined directly, which are then transferred to a second control loop.

The adjustment of the new default values p _k advantageously takes place in a separate control loop. This is advantageous because at any time from the measurements in the first control loop M new default values for the values y _{i can} come out, but only one value for the external parameter p _k can be set at each time.

In this second control loop, a sequence of the settings p _{k is} set. Between two different settings of this sequence, at least one time is waited for, which corresponds to the mechanical inertia of the means for setting the external parameter p. Advantageously, a maximum of 50 times as long is maintained, in particular a maximum of 20 times as long, in particular 5 times as long.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 2007/0152171 A1 [0085]
• DE 10358225 B3 [0085]
• EP 1072957 A2 [0090]
• US 6198793 B1 [0090]
• DE 102013223935 [0093, 0104]
• WO 2009/100856 A1 [0117]

Claims (12)

Radiation source module ( 2 ) for a lighting system ( 19 ) of a projection exposure system ( 1 ) with a plurality M of scanners ( 3 _i ) 1.1. with a coupling-out optics ( 9 ) with at least one means for splitting a collection output beam ( 8th ) having a total intensity in a plurality (M) of single output jets ( 10 _i ) with single intensities, 1.2. wherein the means for splitting the collection output beam ( 8th ) means for varying a division ratio of the total intensity of the collection output beam ( 8th ) into the individual intensities of the single output jets ( 10 _i ). Radiation source module ( 2 ) according to claim 1, characterized by means for periodically varying the division ratio of the total intensity of the collection output beam ( 8th ) into the individual intensities of the single output jets ( 10 _i ), wherein the variation of the division ratio has a period (T _var ) which is at most 10 ms. Radiation source module ( 2 ) according to one of the preceding claims, characterized in that it comprises at least one means for controlling the total intensity of the collecting output beam ( 8th ). Radiation source module ( 2 ) according to one of the preceding claims, characterized in that as means for controlling the total intensity of the collecting output beam ( 8th ) a radiation source ( 4 ) with a controllable intensity of an illumination radiation emitted by the latter ( 5 ), wherein the intensity of the radiation source ( 4 ) emitted illumination radiation ( 5 ) is controllable with a frequency of at least 10 kHz. Radiation source module ( 2 ) according to one of claims 3 to 4, characterized in that the means for controlling the overall intensity of the collecting output beam ( 8th ) depending on the means for dividing the collection output beam ( 8th ) in single output jets ( 10 _i ) is controllable. Radiation source module ( 2 ) according to one of claims 1 to 5, characterized in that a proportion of at least 50% of the single output jets ( 10 _i ) during at least 50% of a period (T _var ) of the variation of the execution ratio of the total intensity of the collection output beam ( 8th ) into the individual intensities of the single output jets ( 10 _i ) have individual intensities which in each case amount to at least 10% of the mean individual intensity of all individual output beams ( 10 _i ) amount. Lighting system ( 19 ) for a projection exposure system ( 1 ) with a plurality of scanners ( 3 _i ) comprising 7.1. a radiation source module ( 2 ) according to any one of claims 1 to 6 and 7.2. a plurality (M) of beam guiding optics ( 13 _i ) for guiding the individual output jets ( 10 _i ) to different object fields ( 11 _i ). Method for controlling the distribution of one of a common radiation source ( 4 ) emitted illumination radiation ( 5 ) on a plurality of scanners ( 3 _i ) comprising the following steps: 8.1. Providing a lighting system ( 19 ) according to claim 7, 8.2. Specifying target intensities of the single output jets ( 10 _i ), 8.3. Determining a time profile of the distribution ratio of the total intensity of the collection output beam ( 8th ) into the individual intensities of the single output jets ( 10 _i ) depending on a periodically variable setting of the means for varying this division ratio, 8.4. Determining the average actual intensities of each of the single output jets ( 10 _i ), 8.5. Determining correction factors for adjusting the average actual intensities of each of the single output jets ( 10 _i ) to the specified target intensities, 8.6. Controlling the total intensity of the collection output beam ( 8th ) depending on the correction factors and on the periodic variation of the distribution ratio variation means. Method for controlling the distribution of one of a common radiation source ( 4 ) emitted illumination radiation ( 5 ) on a plurality of scanners ( 3 _i ) comprising the following steps: 9.1. Providing a lighting system ( 19 ) according to claim 7, 9.2. Specifying target intensities of the single output jets ( 10 _i ), 9.3. Determining the division ratio of the total intensity of the collection output beam ( 8th ) into the individual intensities of the single output jets ( 10 _i ) depending on a variable setting of the means for varying this split ratio, 9.4. Determining a temporal sequence p _k of settings of the distribution ratio variation means as a function of the predetermined target intensities, by a desired distribution ratio of the total intensity of the collection output radiation (p) averaged over the sequence p _k ( 8th ) on the single output jets ( 10 _i ) 9.5. Determining the actual intensities of each of the single output jets ( 10 _i ) when the sequence p _{k of} the settings of the means for varying the distribution ratio, 9.6. Determining deviations of the determined actual intensities of the single output jets ( 10 _i ) from the specified target intensities, 9.7. Adjust the overall intensity of the collection output beam ( 8th ) and / or the sequence of the settings p _k as a function of the deviations. Projection exposure system ( 1 ) for projection lithography comprising 10.1. a lighting system ( 19 ) according to claim 7 and 10.2. a plurality of scanners ( 3 _i ). Method for producing a micro- or nanostructured component comprising the following steps 11.1. Provision of a projection exposure system ( 1 ) according to claim 10, 11.2. Provision of at least one reticle ( 22 _i ), 11.3. Provision of at least one wafer ( 25 _i ) with one for the illumination radiation ( 5 ) sensitive coating, 11.4. Projecting at least a portion of the at least one reticle ( 22 _i ) on the at least one wafer ( 25 _i ) using the projection exposure system ( 1 ), 11.5. Develop the with the illumination radiation ( 5 ) exposed photosensitive layer on the wafer ( 25 _i ). Component produced by the method according to claim 11.

Images