Method for loading a blank composed of fused silica with hydrogen, lens element and projection lens

Abstract

A method for loading a blank composed of fused silica with hydrogen, including loading the blank at a first temperature (T.sub.1) and a first hydrogen partial pressure (p.sub.1), and further loading the blank at a second temperature (T.sub.2) which is different from the first temperature and at a second hydrogen partial pressure (p.sub.2) which is different from the first hydrogen partial pressure. The first and second temperatures (T.sub.1, T.sub.2) are lower than a limit temperature (T.sub.L) at which a thermal formation of silane in the fused silica of the blank commences. Also disclosed are a lens element produced from such a blank and a projection lens that includes at least one such lens element.

Claims

1. A method for loading a blank composed of fused silica with hydrogen, comprising: loading the blank at a first temperature (T.sub.1) of less than 475 C. and a first hydrogen partial pressure (p.sub.1), and further loading the blank at a second temperature (T.sub.2) of less than 300 C. and at a second hydrogen partial pressure (p.sub.2) which is different from the first hydrogen partial pressure, wherein the first temperature and the second temperatures (T.sub.1, T.sub.2) are lower than a limit temperature (T.sub.L) at which a thermal formation of silane in the fused silica of the blank commences, and wherein the loading and the further loading results in a silane content in all locations of the blank of less than 510.sup.14 molecules/cm.sup.3.

2. The method according to claim 1, wherein the further loading of the blank is at a higher hydrogen partial pressure (p.sub.2>p.sub.1) than is the loading.

3. The method according to claim 1, wherein the limit temperature (T.sub.L) at which the thermal formation of the silane in the fused silica of the blank commences is determined prior to the loading.

4. The method according to claim 3, wherein the limit temperature (T.sub.L) is determined based on a change (dK/dF) in absorption coefficient (K) of the fused silica of the blank with respect to an energy density (F) during irradiation of the blank with pulsed laser radiation at a wavelength of 193 nm.

5. The method according to claim 1, wherein the blank is further loaded at a temperature (T.sub.2) of less than 400 C.

6. The method according to claim 1, wherein the blank is loaded at a hydrogen partial pressure (p.sub.1) of less than 20%.

7. The method according to claim 1, wherein the blank is further loaded at a hydrogen partial pressure (p.sub.2) of 50% or more.

8. A lens element composed of fused silica and providing a surface and a further surface configured to transmit radiation therethrough, wherein, at a distance (d.sub.1) of 1 mm from the surface, the lens element has a hydrogen content ([H.sub.2]) of at least 510.sup.16 molecules/cm.sup.3, wherein, in an inner volume of the lens element, the hydrogen content ([H.sub.2]) falls to a minimum value of 10% or less of the hydrogen content ([H.sub.2]) at the surface, and wherein, at a distance (d.sub.2) of 8 mm from the surface, the hydrogen content ([H.sub.2]) falls to 50% or less of the hydrogen content ([H.sub.2]) at the surface; and wherein at all locations in the volume of the lens element a silane content is less than 510.sup.14 molecules/cm.sup.3.

9. The lens element according to claim 8, wherein a silane content varies over a thickness (D) of the lens element by not more than 50%.

10. The lens element according to claim 8, wherein, at a distance (d.sub.2) of 8 mm from the surface, the hydrogen content ([H.sub.2]) falls to 40% or less of the hydrogen content ([H.sub.2]) at the surface.

11. The lens element according to claim 8, wherein the hydrogen content ([H.sub.2]) as far as a distance (d.sub.3) of 25 mm from the surface falls to 10% or less of the hydrogen content ([H.sub.2]) at the surface.

12. The lens element according to claim 8, wherein the hydrogen content ([H.sub.2]) rises in a thickness direction (z) of the lens element from the minimum value to the further surface for passage of radiation to not more than 30% of the hydrogen content ([H.sub.2]) at the surface.

13. The lens element according to claim 8, which has, during irradiation with pulsed laser radiation at a wavelength of 193 nm, a change dK/dF in absorption coefficient (K) with respect to an energy density (F) of less than 510.sup.4 cmpulse/mJ.

14. The lens element according to claim 8, which has an OH content of less than 200 ppm by weight.

15. The lens element according to claim 8, which has a refractive index (n) of less than 1.560830 for radiation at a wavelength (.sub.B) of 193.368 nm.

16. The lens element according to claim 8, wherein a wavefront distortion resulting from compaction during an irradiation of the lens element with pulsed laser radiation at a wavelength of 193 nm rises no more than linearly with pulse number of the laser radiation.

17. The lens element according to claim 8, which exhibits no formation of microchannels after receiving a dose of 200 billion pulses having a pulse duration of in each case 150 ns at an energy density of 0.5 mJ/cm.sup.2/pulse or at a dose of 3.5 billion pulses having a pulse duration of in each case 130 ns at an energy density of 6.5 mJ/cm.sup.2/pulse.

18. A projection lens, comprising at least one lens element according to claim 8.

19. The projection lens according to claim 18, wherein the lens element is a last lens element of the projection lens.

20. A lens element composed of fused silica and providing a surface and a further surface configured to transmit radiation therethrough, wherein at least at one location in an inner volume of the lens element a hydrogen content ([H.sub.2]) is less than 210.sup.15 molecules/cm.sup.3, wherein the hydrogen content ([H.sub.2]) at the surface and at the further surface is between 1.5 times and 5 times the hydrogen content ([H.sub.2]) at the at least one location, wherein a silane content at the surface and at the further surface of the lens element is between 1.1 times and 2.20 times the silane content at the at least one location; and wherein at all locations in the volume of the lens element a silane content is less than 510.sup.14 molecules/cm.sup.3.

21. A lens element composed of fused silica, wherein a ratio between a maximum silane content and a minimum silane content in a volume of the lens element is lower than a square root of the ratio between a maximum hydrogen content ([H.sub.2]) and a minimum hydrogen content ([H.sub.2]) in the volume of the lens element and wherein at all locations in the volume of the lens element a silane content is less than 510.sup.14 molecules/cm.sup.3.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Exemplary embodiments are illustrated in the schematic drawing and are explained in the following description. In the figures:

(2) FIG. 1 shows a schematic illustration of one exemplary embodiment of a projection exposure apparatus for immersion lithography,

(3) FIG. 2 shows a schematic illustration of a last lens element of a projection lens of the projection exposure apparatus from FIG. 1,

(4) FIG. 3 shows a schematic illustration of the distribution of the energy density in the thickness direction that occurs in the last lens element,

(5) FIG. 4 shows a schematic illustration of the distribution of the hydrogen content and of the silane content of the last lens element in the thickness direction,

(6) FIG. 5 shows an illustration analogous to FIG. 4 for a further exemplary embodiment of a last lens element,

(7) FIG. 6 shows one example of the sequence of a two-stage hydrogen loading process for the lens elements in accordance with FIG. 4 and FIG. 5, and

(8) FIG. 7 shows an illustration of a further lens element for the projection lens from FIG. 1.

DETAILED DESCRIPTION

(9) In the following description of the drawings, identical reference signs are used for identical or functionally identical components.

(10) FIG. 1 schematically shows a microlithographic projection exposure apparatus 1 provided for producing large scale integrated semiconductor components using immersion lithography. The projection exposure apparatus 1 comprises as light source an excimer laser 3 having an operating wavelength of approximately 193 nm. Alternatively, light sources having other operating wavelengths, for example 248 nm or 157 nm, could also be used. An illumination system 5 disposed downstream generates in its exit plane or object plane 7 a large, sharply delimited, very homogeneously illuminated illumination field adapted to the telecentricity requirements of the projection lens 11 disposed downstream. The illumination system 5 has devices for controlling the pupil illumination and for setting a predefined polarization state of the illumination light. In the beam path downstream of the illumination system 5, a device (reticle stage) for holding and moving a mask 13 is arranged such that the latter lies in the object plane 7 of the projection lens 11 and is moveable in said plane for scanning operation in a driving direction 15.

(11) The projection lens 11 follows downstream of the object plane 7, also designated as mask plane, said projection lens imaging an image of the mask with a reduced scale onto a substrate 19, for example a silicon wafer, said substrate being covered with a photoresist, also called resist 21. The substrate 19 is arranged such that the planar substrate surface with the resist 21 substantially coincides with the image plane 23 of the projection lens 11. The substrate is held by a device 17 comprising a drive for moving the substrate 19 synchronously with the mask 13. The device 17 also comprises manipulators for moving the substrate 19 both in the z-direction parallel to the optical axis 25 of the projection lens 11 and in the x- and y-directions perpendicular to said axis.

(12) The device 17 (wafer stage) provided for holding the substrate 19 is designed for use in immersion lithography. It comprises a receptacle device 27, which is moveable by a scanner drive and the base of which has a flat cutout for receiving the substrate 19. By virtue of a circumferential edge 29, a flat, upwardly open, liquid-tight receptacle for an immersion liquid 31 is formed. The height of the edge is dimensioned such that the introduced immersion liquid 31 can completely cover the substrate surface 21 with the resist and the exit-side end region of the projection lens 11 can dip into the immersion liquid 31 with a correctly set working distance between lens exit and substrate surface 21. The immersion liquid 31 can also be introduced between the lens exit and the resist 21 in a different manner from that illustrated in FIG. 1.

(13) The projection lens 11 has an image-side numerical aperture NA of at least NA=1.2, typically of more than NA=1.3, and is thus particularly adapted to use with immersion liquids. The projection lens 11 has an approximately hemispherical planoconvex lens 33 as the last imaging optical element closest to the image plane 23, the exit surface 35 of said planoconvex lens element forming the last optical surface of the projection lens 11. The exit side of the last lens element 33 is completely immersed in the immersion liquid 31 during the operation of the projection exposure apparatus 1 and is wetted by said liquid. The hemispherical planoconvex lens element 33 consists of fused silica in the same way as a further lens element 37 of the projection lens 11 which is illustrated by way of example in FIG. 1.

(14) The fused silica of the last lens element 33 is a dry fused silica having an OH content of typically less than 10 ppm (by weight), in particular of less than 5 ppm or of less than 1 ppm (by weight), which was produced in a soot process. The fused silica of the last lens element 33 has been annealed to a low defect density i.e. its fictive temperature T.sub.f is below 1000 C., its refractive index n at a vacuum wavelength .sub.B of 193.368 nm, which corresponds to the wavelength of the laser 3 used for generating the laser radiation 4, is between approximately 1.560795 and 1.560830 at room temperature (22 C.). While the calibration of the fictive temperature T.sub.f is not uniform, the refractive index n, to put it more precisely the real part thereof, can be detected very accurately. For the same OH content of the fused silica, a lower refractive index n of the fused silica means a lower fictive temperature T.sub.f.

(15) During the operation of the projection exposure apparatus 1, the fused silica material of the lens elements 33, 37 is irradiated by the excimer laser 3 (ArF laser) with intensive laser pulses at the operating wavelength of 193 nm. Depending on the choice of the illumination settings (e.g. in the case of dipole or quadrupole illumination), in particular the last lens element 33, illustrated in FIG. 2, in partial regions of its volume, in particular in direct proximity to the exit surface 35, is subjected to laser radiation 4 having high pulse energy densities of, for example, approximately 0.5 mJ/cm.sup.2 or higher, the pulse duration typically being approximately 150 ns or higher. The pulse duration is understood to meanas generally customarythe equivalent pulse duration or the so-called total integral square (TIS), i.e. the ratio of the square of the time integral over the intensity of a pulse to the time integral over the squared intensity of said pulse.

(16) As can be discerned in FIG. 3, in the volume of the last lens element 33 in the thickness direction z of the lens element 33 the energy density F decreases continuously starting from the light exit surface 35 to the opposite light entrance surface 36. At the location of half the thickness D/2=2.5 cm of the last lens element 33, the energy density F is only approximately 0.1 mJ/cm.sup.2/pulse, i.e. it is lower than at the exit surface 35 by a factor of five. The curve shown in FIG. 3 shows, for each position in the thickness direction z, the maximum energy density in a plane (i.e. in the x,y-direction) of the lens element 33 that is assigned to said position. In the example shown, the last lens element 33 has a thickness D of 5 cm and the thickness of a blank 32 from which the last lens element 33 was produced is slightly larger since said blank has a (small) oversize.

(17) The last lens element 33 is designed for a load of approximately 200 billion pulses having the energy density F respectively shown in FIG. 3 before replacement becomes necessary. In order to achieve this lifetime, in the case of the above-described energy density F at the exit surface 35 or directly adjacent to the exit surface 35, i.e. as far as a distance d.sub.1 1 mm from the exit surface, a hydrogen content of approximately 110.sup.17 molecules/cm.sup.3 is required, as is evident with reference to the distribution curve in FIG. 4. This value for the maximum hydrogen content is an exemplary value; typical values for the hydrogen content (or the deuterium content) are between approximately 510.sup.16 molecules/cm.sup.3 and 110.sup.18 molecules/cm.sup.3.

(18) Typically, the required hydrogen content in a respective volume region of a lens element composed of fused silica depends approximately quadratically on the energy density occurring there. In the case of an energy density F lower by a factor of five in the centre of the last lens element 33, accordingly only a hydrogen content of approximately 410.sup.15 molecules/cm.sup.3 is required there. For reasons of process reliability (reliable detection with Raman spectroscopy or in the case of omission of measurements after loading buffer for concentration fluctuations caused by temperature fluctuations in the loading furnace and fluctuations of the defect density in the glass), a minimum hydrogen content of 110.sup.16 molecules/cm.sup.3 is defined for the centre of the last lens element 33 (at a (third) distance d.sub.3=2.5 cm from the (planar) exit surface 35) or of the associated blank 32. It has proved to be advantageous if the hydrogen content [H.sub.2] falls comparatively rapidly from the exit surface 35 and, for example at a second distance d.sub.2=8 mm from the exit surface 35, falls to at least 50% of the maximum value of the hydrogen content [H.sub.2] that is present at the exit surface 35. Overall, this therefore results in the distributionadvantageous for the present use and illustrated in FIG. 4of the hydrogen content [H.sub.2] in the thickness direction (z-direction) of the last lens element 33.

(19) In order to produce the distribution of the hydrogen content [H.sub.2] in the last lens element 33 as shown in FIG. 4, a hydrogen loading of the blank 32 from which the last lens element 33 is produced is performed, which is described in greater detail below with reference to FIG. 6. For loading, the blank 32 is introduced into a furnace (not shown), in which a hydrogen-containing atmosphere prevails and which is heated to a constant temperature during conventional loading in order that the hydrogen from the surroundings can diffuse into the blank 33. The duration of the loading with hydrogen depends on the desired hydrogen content and on the chosen temperature during the loading with hydrogen. Typically, the loading takes a number of weeks or months, wherein the loading can possibly take a year or longer at low temperatures of e.g. approximately 300 C.

(20) In order to shorten the loading duration and nevertheless ensure that not too much silane (SiH) forms in the fused silica material, it is proposed to modify the known loading process as follows: The loading of the blank 32 is carried out in two loading stages (cf. FIG. 6), wherein in the first loading stage the blank 32 is loaded at a first temperature T.sub.1 and a first hydrogen partial pressure p.sub.1, and in the second loading stage the blank 32 is loaded at a second temperature T.sub.2, which is different from the first temperature, and a second hydrogen partial pressure p.sub.2 which is different from the first hydrogen partial pressure.

(21) Typically, in this case the first loading stage is carried out at a first hydrogen partial pressure p.sub.1 that is lower than the second hydrogen partial pressure p.sub.2 in the second loading stage. Moreover, the temperature T.sub.1 in the first loading stage is generally chosen to be greater than the temperature T.sub.2 in the second loading stage. In this way, for example, the hydrogen distribution in the blank 32 as shown in FIG. 4 can be produced, as is explained in detail below:

(22) In the first stage, a value of 450 C. is chosen for the loading temperature T.sub.1. At 100% hydrogen partial pressure at a total pressure of 1 atm in the loading furnace, a concentration or a hydrogen content of approximately 310.sup.17 molecules/cm.sup.3 is established in the equilibrium state in the fused silica. Accordingly, with infinitely long loading at this temperature T.sub.1, a hydrogen concentration of approximately 3% at standard pressure would suffice to achieve the required value of approximately 110.sup.16 molecules/cm.sup.3 in the centre of the blank 32. For loading times of the order of magnitude of a few weeks or a few months, a partial pressure p.sub.1 of approximately 10%, for example, will be chosen in the first stage. As a result, a hydrogen content of approximately 310.sup.16 molecules/cm.sup.3 is established in the outer regions of the blank 32, i.e. in particular adjacent to the later exit surface 35 of the last lens element 33, which still does not lead to relevant SiH formation at the chosen temperature T.sub.1 of approximately 450 C. in the first loading stage. The first loading duration t.sub.1t.sub.0 up to the point in time t.sub.1 can be chosen to be short enough that only approximately 80-90% of the desired content of 110.sup.16 molecules/cm.sup.3 is established in the centre of the blank 32 since the hydrogen introduced in the first loading stage diffuses further inwards during the second loading stage (from t.sub.1 to t.sub.2) and, as a result, the desired value is attained in the centre of the blank 32. The first loading stage lasts approximately as long as is necessary for an averagely or slightly burdened lens element position having the same thickness at the same loading temperature T.sub.1 i.e. the loading duration (t.sub.1t.sub.0) of the first stage is approximately 10 weeks.

(23) In the second stage, the hydrogen required in the outer regions of the blank 32 is introduced at a significantly lower temperature T.sub.2, for example at less than approximately 300 C., e.g. at approximately 250 C. A hydrogen partial pressure p.sub.2 of 50% relative to standard pressure is chosen in the second stage. This has the effect that the minimum content of 110.sup.17 molecules/cm.sup.3 is reliably attained in the outermost regions of the blank 32. Since the solubility of hydrogen is higher at low temperatures, a higher content of approximately 210.sup.17 molecules/cm.sup.3 can possibly also be established there. As was explained further above, at a distance d.sub.2 of 8 mm from the exit surface 35 the hydrogen content is permitted to have already fallen to half (50%), such that at the specified temperature a loading duration (t.sub.2t.sub.1) in the second loading stage of a few weeks is sufficient.

(24) A pressure in the loading furnace or autoclave of higher than standard pressure (1 atm) is permitted to be chosen in particular for the second loading stage. Likewise, the temperature is permitted to fall to room temperature (22 C.) again between the first and second stages, since no relevant diffusion of hydrogen into the blank 32 takes place any longer at room temperature. This enables blanks to be collected and stored, which blanks can then be subjected to the second loading stage in the furnace in a common batch.

(25) The temperatures T.sub.1, T.sub.2 and partial pressures p.sub.1, p.sub.2 specified here should be understood to be merely by way of example. The partial pressures p.sub.1, p.sub.2 are typically coordinated with one another such that the desired hydrogen content is established in the centre of the blank 32 in an economically tenable total process duration (t.sub.0 to t.sub.2). Depending on the application, the first partial pressure p.sub.1 can be for example approximately 20% or less, and the second partial pressure p.sub.2 approximately 50% or higher, if appropriate 90% or higher, in particular even 100% or higher (if a pressure above standard pressure, i.e. of more than 1 atm, prevails in the furnace). In particular, the first temperature T.sub.1 can be between approximately 400 C. and 500 C., but temperatures that are significantly higher are also possible, as is described in greater detail further below. The second temperature T.sub.2 is typically 400 C. or lower, but can in particular also be 300 C. or less.

(26) In order to define the loading temperature T.sub.1 in the first stage, the following procedure can be adopted: Firstly, a hydrogen content appropriate for the later application is chosen for a basic glass (prior to loading), said hydrogen content being determined by the maximum expected energy density and pulse number on the respective lens element in the optical system. Afterwards, samples or blanks loaded at different temperatures are produced and subjected to an FDT test, for example as explained in U.S. Pat. No. 7,928,026 B2, and a limit temperature T.sub.L is found starting from which the FDT value, i.e. the gradient dK/dF of the absorption coefficient K (cf. FIG. 2), rises significantly (for the glasses from FIG. 2 of U.S. Pat. No. 7,928,026 B2 for example T.sub.L of approximately 475 C.). In particular, the limit temperature T.sub.L can also be determined by a comparison of the gradient dK/dF with a threshold value, which can be for example 510.sup.4 cmpulse/mJ or lower. A lower temperature T.sub.1 with a sufficient separation from the point starting from which the gradient increases significantly (i.e. from the limit temperature T.sub.L) is defined for the actual loading in series production. The loading temperature T.sub.2 in the second stage is also chosen to be lower than the limit temperature T.sub.L.

(27) Since the FDT gradient is only one necessary indicator of good lifetime behaviour, for complete qualification the fused silica material can additionally be subjected to a long-term test in which it is irradiated for billions of pulses with energy densities of up to approximately one order of magnitude above the later use conditions and is then examined for compaction, induced absorption, hydrogen consumption, etc. It is also possible to choose different loading temperatures for lens element positions which are subjected to loads of different magnitudes and which require a different hydrogen content and different laser durability, with the same basic glass.

(28) For a given (first) loading temperature T.sub.1, it should now also be clarified how the SiH concentration is dependent on the hydrogen concentration. Generally, the type of precursor defects of the basic glass will also influence this relationship, but starting points for this relationship can be derived with reference to Table 1 of U.S. Pat. No. 7,928,026 B2: For pairs of blanks having the same loading temperature and approximately the same OH content and the same fictive temperature, the FDT gradient (dK/dF) behaves predominantly like the square root of the hydrogen contents. With this relation being taken as a basis, in the case of conventional loading at a constant temperature in the case of the hydrogen distribution shown in FIG. 4 in the outer regions of the blank 32 a high silane content would be established here, specifically typically a ratio of 3.1:1 in the silane content in the case of the ratioillustrated in FIG. 4of 10:1 in the case of the hydrogen content [H.sub.2] from the exit surface 35 to the centre of the blank 32.

(29) In the case of the two-stage method proposed further above, there results only a small increase in the silane content in the outer regions of the blank, which is of the order of magnitude of a few 10% in the region of the exit surface 35, as can likewise be discerned in FIG. 4, which shows the silane content [SiH] (in arbitrary units), said silane content typically being of an order of magnitude of between approximately 110.sup.13 molecules/cm.sup.3 and 510.sup.15 molecules/cm.sup.3, wherein ideally at all locations in the volume of the lens element or of the blank 32 the silane content is less than 510.sup.14 molecules/cm.sup.3, less than 110.sup.14 molecules/cm.sup.3, or less than 510.sup.13 molecules/cm.sup.3. The silane content thus varies over the volume of the blank 32 (in this thickness direction z) by less than 30%, that is to say that the (minimum) value in the centre of the blank 32 is less than 30% below the maximum value of the silane content which occurs at the edge of the blank 32 in the region of the later exit surface 35.

(30) Since Raman spectroscopy, conventionally used for determining the silane content, generally has an excessively high detection limit for lithography applications, the FDT gradient can be used as a measure of the silane content, said gradient being defined as the change dK/dF in the absorption coefficient K of the fused silica depending on the energy density F during irradiation with pulsed laser radiation at the operating wavelength .sub.B=approximately 193 nm used here. In the case illustrated above, there results an FDT gradient (dK/dF) of less than 510.sup.4 cmpulse/mJ, of less than 110.sup.4 cmpulse/mJ, of less than 510.sup.5 cmpulse/mJ, ideally of less than 110.sup.5 cmpulse/mJ, in conjunction with a short and hence economic loading duration.

(31) Overall, the loading method described here makes it possible to obtain a profile of the hydrogen content which is tailored to the requirement profile of the last lens element 33 and which would not have been possible in the case of conventional loading at a uniform loading temperature: The high hydrogen content is very closely restricted to the outer millimeters of the lens element 33 or of the blank 32 and does not have the very excessive increase that results in the case of loading at more than standard pressure. The refractive index profile in the lens element 33, said profile being dependent on the hydrogen content, is flatter and more reproducible and said lens element is therefore handleable more easily. As a result of the loading of the outer regions at a lower temperature T.sub.2, the silane content is increased only insignificantly, as a result of which a better performance over the lifetime and a higher loading according to the full lifetime striven for are possible. In the inner portion of the last lens element 33, the actual curveshown in FIG. 4of the hydrogen content [H.sub.2] lies only just above the requirement curve (not illustrated), as a result of which less absorption and less change in absorption as a function of the energy density arise upon integration along the beam path through the lens element 33.

(32) If the mirrored profile of the hydrogen content [H.sub.2] on the other side of the blank 32 or at the light entrance surface 36 of the last lens element 33 proves to be disturbing, the blank 32, as illustrated in FIG. 5, can be produced for example with an oversize of e.g. 1 to 2 cm thickness and this oversize can be separated after loading, as is indicated in FIG. 5. As is evident with reference to FIG. 5, the hydrogen content [H.sub.2] rises only insignificantly in the thickness direction (z-direction) from the minimum value, which is assumed in the centre of the blank 32, as a result of the separation of the oversize up to the light entrance surface 36, where it reaches not more than 30% of the content of hydrogen at the light exit surface 35, that is to say that it is possible in this way to produce a non-mirror-symmetrical hydrogen distribution in the last lens element 33.

(33) The above-described method for loading a blank with hydrogen can be used not just in the case of lens elements which are impinged on by radiation having a high energy density and accordingly have to be loaded with a high hydrogen content; rather, the method described above can advantageously be used for loading arbitrary lens elements, particularly in the case of lens elements having a low load and/or a small difference in the energy density on the front and rear sides, as explained below by way of example on the basis of the concavo-convex meniscus lens element 37 illustrated in FIG. 7. As is indicated in FIG. 7, the meniscus lens element 37 is arranged in the convergent beam path of the laser radiation 4, that is to say that the convexly curved lens element surface 36 constitutes the light entrance surface and the concavely curved lens element surface 35 constitutes the light exit surface, wherein, as described above, the differences in the energy densities at the two surfaces 35, 36 are comparatively small.

(34) In the present exemplary embodiment, it is proposed to determine the density of reactive defects and contaminants of the basic glass (i.e. of the glass prior to loading) theoretically or experimentally and to define a mean content in a blank 38, which is composed of the maximum defect density, the requirement calculated from the energy density in the centre, and a safety factor, and is typically of the order of magnitude of 0.2 to 210.sup.15 molecules/cm.sup.3, i.e. is lower than the hydrogen content usually set in the centre of the blank 38, said hydrogen content being approximately 5 to 1010.sup.15 molecules/cm.sup.3.

(35) In the first stage of loading the blank 38, a partial pressure p.sub.1 of typically between 0.25% and 2.5% is chosen, which, given a suitably chosen first loading duration t.sub.1t.sub.0 of a number of weeks or months, leads to a hydrogen content at the edge of the blank 38 which is 1.5 to 5 above the hydrogen content in the centre of the blank 38, i.e. typically as far as a distance of 1 mm from the respective surface 35, 36 of the meniscus lens element 37 leads to a hydrogen content of between approximately 0.310.sup.15 molecules/cm.sup.3 and 110.sup.16 molecules/cm.sup.3.

(36) In this example, the first stage of loading is carried out at a markedly high temperature T.sub.1 of generally between approximately 500 C. and 1000 C., preferably between 700 C. and 1000 C., wherein care should be taken to ensure that a fused silica material is used which has a limit temperature T.sub.L that is above the first temperature T.sub.1. At such high temperatures, temperatures that deviate by e.g. 25 C. during the loading are scarcely of any significance. Since the hydrogen that diffuses into the fused silica material still reacts locally with defects present there, high concentrations of free hydrogen do not occur, and so an excessively high silane content is not established in the fused silica blank 38 either.

(37) In the second stage of loading, the increased hydrogen content required in the edge regions or at the surfaces 35, 36 is introduced at a lower temperature T.sub.2, for example between 400 C. and 500 C., and a higher partial pressure p.sub.2 of between approximately 1% and 20%, the duration t.sub.2t.sub.1 of the second loading stage being approximately six weeks, for example. Besides the low total content of silane, which, detected using the FDT gradient dK/dF, in this example, too, is less than 510.sup.4 cmpulse/mJ, on account of the two-stage loading the silane content at the surfaces 35, 36 or at the outer sides of the blank 38 is between 1.1 times and 2.20 times the silane content in the centre of the blank 38. The ratio of maximum to minimum silane content is thus lower than the square root of the ratio between the maximum hydrogen content and the minimum hydrogen content in the volume of the blank 38.

(38) In the variant of the method described here, the target content in the centre of the blank 38 after the first stage is calculated only from the defect density and a safety allowance. The second loading stage corresponds to the previously known single-stage loading process at T.sub.2=400-500 C., but it is possible to dispense with the overloading in the centre of the blank 38 owing to an unknown defect density, which significantly shortens the loading process overall. This is particularly advantageous if the thickness of the blank 36, 38 is comparatively large and is e.g. more than approximately 5 cm, since loading durations of more than a year can result in this case in a conventional loading method at a constant temperature of e.g. 300 C.

(39) During a loading stage at a constant temperature T.sub.1 or T.sub.2, if appropriate, the hydrogen partial pressure p.sub.1, p.sub.2 can be varied over time, as is explained in U.S. Pat. No. 6,810,687 B2 or 7,994,083 B2 cited in the introduction. Particularly in the second stage, a total pressure of more than 1 atm can also be set in the furnace. As can likewise be discerned with reference to FIG. 7, a three-dimensional distribution of the hydrogen content is established in the blank 38, in the case of which distribution areas of identical hydrogen concentration form ellipsoids of revolution, a minimum hydrogen content being established in the centre of the cylindrical blank 38. In the case shown, the concave lens element surface 35 is comparatively far away from the end side of the blank 38, such that there the hydrogen content is possibly significantly lower than at the end side of the blank 38. In the case of such a meniscus lens element 37, it may therefore be advantageous to carry out mechanical preprocessing prior to loading with hydrogen in order to adapt the geometry of the blank 38 to the later lens element geometry, as is described for example in DE 10 2007 022 881 A1, which is incorporated by reference in the content of this application.

(40) With the aid of the method described above, it is possible to implement temporally practicable loading in the second stage and, if appropriate, also in the first stage at temperatures below 400 C. or below 300 C. particularly in the case of comparatively small blanks. In this way, it can be ensured that a wavefront distortion of the lens elements 33, 37 on account of compaction during irradiation with pulsed laser radiation at a wavelength of 193 nm does not rise more than linearly with the pulse number of the pulsed laser radiation, and that no microchannels form at the lens elements 33, 37 even after irradiation with approximately 200 billion pulses at energy densities of 0.5 mJ/pulse, i.e. over their entire lifetime.