Open Access
Add to BibSonomyAbstract
The influence on Nickel-like Molybdenum soft-x-ray laser performance and stability of a low energy laser prepulse arriving prior to the main laser pumping pulses is experimentally investigated. A promising regime for 10 Hz operation has been observed. A four times increase in soft-x-ray laser operation time with a same target surface is demonstrated. This soft-x-ray laser operation mode corresponds to an optimum delay between the prepulse and the main pulses and to a prepulse energy greater than 20 mJ. We also show that this regime is not associated with a weaker degradation of the target or any reduced ablation rate. Therefore the role of preplasma density gradient in this effect is discussed.
© 2012 Optical Society of America
1. Introduction
With rapid progress in pumping efficiency, saturation operation of plasma-based soft-x-ray lasers can be obtained on low energy laser systems operating at high repetition rate of 10 Hz or more [1]. True 10 Hz routine operation is now possible on a time scale limited by the target lifetime or by the driving laser stability. This issue received few attention because most of the applications explored so far were performed in single shot mode, e.g. for pump-probe experiments [2, 3]. However, some emerging applications require the accumulation of hundreds of shots for surface nano-patterning [4], radiation dose deposition in biological samples [5] or scanning of large experimental parameter range [6, 7]. A simple solution is the use of gas target that can be easily refreshed, but the possible soft-x-ray laser lines are in this case limited to wavelength above 20 nm [8, 9]. On the other hand, saturated operation has been demonstrated down to 10 nm [10] using solid target and strong laser emission below this wavelength is reached [11]. In this case, the degradation of the target at each shot will be an issue for long continuous operation. A first solution consists in continuously moving the target. Helicoidal [12] or metal ribbon [13] target have been implemented.
In a previous paper [5] we presented an effective and stable 10 Hz soft-x-ray laser operation at 18.9 nm. This technical development was driven by applications to biological sample irradiation. An improvement of the shot to shot stability was demonstrated due to the soft-x-ray laser pumping configuration where all the laser pulses required to generate the source propagate along the same beam, avoiding any beam overlapping instabilities. A shot to shot energy fluctuation within a 10% standard deviation was reported and the energy integrated over 20000 shots varied by less than 5% from one irradiation sequence to another. In this case laser interaction occured hundreds times at the same place on a slab target before automated target refreshment by a 200 μm vertical translation. It has been observed that the presence of a low energy prepulse obtained by modifying the opening time of the pulse picker after the regenerative amplifier increases the single shot XUV energy but also the total number of shots with a high energy level generated from the same target location.
We performed an extensive study of this phenomenon. A controlled prepulse was purposefully generated and sent on a Molybdenum slab target prior to soft-x-ray laser pumping pulses. The following paper presents the experimental observations in terms of optimum delay, energy ratio and spatial shape for long, stable and efficient soft-x-ray laser operation at 10 Hz.
2. Experimental arrangement
The experiment was performed on the LASERIX facility. The pumping laser is a chirp pulse amplification (CPA) Titanium:Sapphire laser chain delivering pulses of joule energy level, with a 810 nm central wavelength, and a 10 Hz repetition rate. The soft-x-ray laser at 18.9 nm was generated using the double pulse, single beam, grazing incidence pumping configuration or DGRIP [14]. In this experimental arrangement, two optical pulses are generated at the beginning of the laser chain, before the regenerative amplifier. After the pre-amplification they are following the same optical path through the laser chain, compressor and focussing optics. The total energy after compression is 950 mJ. As shown in Fig. 1, both pulses are incident on target with a low grazing incidence angle of 20 degrees. They are horizontaly line-focussed on the target by a spherical mirror of 500 mm focal length used in an off axis configuration [15]. The line focus was slightly longer than the target which is a 4 mm × 50 mm slab of molybdenum.
In this soft-x-ray laser pumping scheme, the first pumping pulse with 200 ps duration, ionises molybdenum to the nickel-like ionisation stage. The second pulse with only 1 ps duration heats rapidly by inverse Bremstrahlung the plasma free electrons. They generate by collisionnal excitation a population inversion on the 4d-4p transition of the nickel-like molybdenum ions. A soft x-ray laser emission at 18.9 nm is generated on the axis of the line-focus by amplification of spontaneous emission. An experimental optimization, led to an energy balance of 25 percents of the total laser energy in the first pulse, an optimum focal line width near 60 micrometer and a delay between the first and second pulse of 150 ps. These parameters correspond to an intensity on target of 5.1011 W/cm2 for the first pulse and 3.1014 W/cm2 for the second one.
The soft-x-ray laser operation at 10 Hz was quantitatively surveyed using a near field XUV imaging device with a high magnification. The optical part is composed of a 500 mm focal length spherical mirror with multilayer Mo:Si treatment to select the specific wavelength of the soft-x-ray laser and to minimize the background coming from the incoherent emission of the plasma. After 6.5 meter propagation under vacuum, the image of the plasma exit is recorded onto a ANDOR back illuminated XUV CCD camera with magnification factor of 13 and a 1 micrometer resolution at the soft-x-ray laser source exit plane.
After the soft-x-ray laser optimization step, a controlled low energy prepulse is introduced and is incident on target prior to the two pumping pulses. For this purpose, 10% of the main beam energy is extracted by a beamsplitter inserted before the final temporal compression. A delay line, a variable attenuator and an additional compressor enable to change the properties of this auxiliary beam which is finally focussed on target at normal incidence with a combination of spherical and cylindrical lenses. For sake of stability, the prepulse line focus width was set to 80 μm × 4.7 mm, which is 25 percent larger than the pumping pulses line focus.
Due to the particular way of generating the two pumping pulses, the temporal structure of what we will refer to the prepulse is eventually composed of two pulses with relative delay and energy ratio identical to those of the main pumping pulses. The auxiliary compressor allows to change the prepulse duration and the attenuator varies the total prepulse energy EPP from 0 to 100 mJ. In the following we call ΔτPP the delay between the two most energetic pulses within the temporal structure of the prepulses and the main pumping pulses. ΔτPP can be changed from 500 ps to 5400 ps using the delay line. The prepulse duration can be modified by changing the distance between the two auxiliary compressor gratings. In the following discussion, we will call τPP the duration of the most energetic prepulse. τPP was set to 32 ps, 88 ps or 188 ps. The corresponding durations for the less energetic prepulse are respectively 212 ps, 262 ps and 368 ps. Note that the delay ΔτPP has been kept constant when the compressor settings were modified by adding or subtracting an optical path correction on the delay line.
As emphasized in the introduction, the evolution of the soft-x-ray laser properties with the number of shots on the same target place is the major issue for 10 Hz operation. During this experiment two procedures have been used to study this aspect. In a first experimental run performed on a restricted set of prepulse parameters, real time near-field movies have been recorded to monitor the true shot to shot evolution over hundreds of shots. The spatial resolution was low due to a CCD binning applied to reduce the camera read-out time below 100 ms. In a second experimental run, images with higher resolution of the soft-x-ray laser were acquired at different moment of the target degradation. After a serie of 30 shots at 10 Hz repetition rate, 5 consecutive single shots were performed and for each of them high resolution images were acquired. This 35 shots cycle was repeated until the x-ray laser signal became lower than the plasma self emission. The four graphs presented in Fig. 2 result from this sampling procedure. A point represent the x-ray laser energy level for a given number of shots already performed on the target. The energy value is the average energy extracted from the five consecutive near field images. The error bar is representative of the typical standard deviation.
Fig. 2 Evolution of the soft-x-ray laser output energy as a function of the number of shots on the same target position.
For the same experimental conditions, the comparison between the two procedure gives identical results for the soft-x-ray laser energy and source position evolution. It has also been checked that the trends presented below were reproducible when a new fresh target position was used.
3. Experimental results
The delay between the prepulse and the main pulses appeared to be the critical parameter to achieve what we will call later a stable operation. Always starting with a fresh target surface, the experimental series presented in Fig. 2 were recorded without any prepulse (first graph) or with different prepulse delay values, ΔτPP = 500 ps, 1.4 ns, 2.75 ns and 5.2 ns. The prepulse energy was set to the maximum achievable value, EPP = 87 mJ and to a same compression (τPP = 32 ps).
In the absence of prepulse, the energy decreased to a negligible value after 60 shots on the same target place. Adding a 87 mJ prepulse with a delay greater than 2 ns led to a similar general evolution: the initial signal level was equivalent and there was a strong decrease of the average energy only after 40 to 60 shots. However, it should be noted that contrary to the situation without prepulse, low energy lasing was sometimes reappearing during one or two shots even after hundreds of shots on target. But these low lasing events were appearing randomly and with so rare occurence that the average procedure makes this contribution negligible. At shorter delays, we observed that both initial signal level and number of soft-x-ray laser shots achievable with a same target location are increased. A two times increase in initial signal level compared to the situation without prepulse is even observed at 0.5 ns. For an optimal delay of 1.4 ns, the soft-x-ray laser energy is almost constant during the first hundred shots and is divided by two after 200 shots. Lasing action is still visible above plasma self emission after 300 shots.
For this delay value, the effect of the prepulse total energy EPP has been investigated. Results are presented in Fig. 3. For EPP > 20 mJ, evolution of soft-x-ray laser energy with shots on target have similar behavior, dramatically different from the situation without prepulse. The initial energy level is ∼ 50% higher and the number of shots that can be performed on the same target position is increased by a factor of 3.
Fig. 3 Evolution of the soft-x-ray laser output energy as a function of the number of shots on the same target position.
In order to give a quantitative and synthetic assessment of each evolution, experimental results were fitted with the f function described below and represented by a solid line on the graphs.
In this formula, n > 0 is the number of shot on the same target position and E1, ntarget and λ are fitting parameters. This function efficiently fits the observed evolutions and help us to determine three characteristic features of the signal evolution: the initial energy level, the lambda parameter and the target lifetime.
The initial energy level f (n = 1) represents the soft-x-ray laser output energy for a fresh target position. The parameter λ is reflecting how flat is the beginning of the energy evolution. If λ is high, the evolution is close to an exponential decay starting at n=1. An exemple of such an evolution can be seen in Fig. 2 for ΔPP = 500 ps. If λ is small, the evolution starts by a plateau of constant value close to E1 followed by a fast decay to zero. In this case, the point n = ntarget correspond to the number of shots after which the soft-x-ray laser energy is divided by two or the position of this fast decay. For this reason ntarget will be called the target lifetime. In summary, efficient and long operations would corresponds to high values of E1 and ntarget. If energy stability is also required during the target lifetime, high values of λ are requested.
The evolution of these four parameters with the prepulse energy EPP and the prepulse compression τPP are summarized in Fig. 4. Another interesting figure of merit is the total soft-x-ray laser energy per target position. This value is of great help to estimate the target requirement for long soft-x-ray laser operation or for experiment requiring XUV integrated dose (nano patterning, radiobiology...). Figure 4d shows the total energy per target position as a function of prepulse energy and compression.
Fig. 4 Evolution of the initial energy (a), target lifetime (b), lambda parameter(c), and integrated dose (d) with the prepulse energy Epp and prepulse compression τPP at a 1.4 ns delay.
A prepulse energy of EPP = 10 mJ as little effect on the target lifetime, the λ parameter and on the integrated dose. However, above this value important changes are observed. Configurations with a pulse compression of τPP = 32 ps and 82 ps lead to a similar behavior. All the parameters and also the initial soft-x-ray laser energy level are dramatically increased when EPP rises from 10 mJ to 20 mJ then reaches a plateau for greater prepulse energy. For an optimal set of parameters (EPP = 50 mJ – τPP = 82 ps) the target lifetime and integrated dose can be increased as high as four times compared to the situation without prepulse. However the behavior for a longer prepulse with τPP = 188 ps is different. The initial energy level is not significantly increased compared to the situation without prepulse but the target lifetime and λ increase slowly with EPP, leading finally to an increase of the total integrated dose.
These trends are reproducible from one target location to another provided that two consecutive locations are separated by at least 200μm. They are also reproducible from one day to another (including restart and realignement of the laser chain) with quantitative differences below 20 percents.
4. Discussion
Introduction of small energy prepulse in soft-x-ray laser pumping scheme is known to increase the performance of the soft-x-ray laser in quasi-steady state (QSS) [18, 22] or transient GRIP configuration [19, 20] and is implemented on versatile soft x-ray laser systems [21]. This aspect has been recently revisited in a pumping configuration close to the one described here [16]. However, this is the first time to our knowledge that this kind of pumping arrangement is shown to improve the target lifetime. It should be noted that this effect is not a direct and trivial consequence of an increase in soft-x-ray laser energy. The most striking illustration is the evolution observed for a prepulse compression of τpp=188 ps: the target lifetime is increased by a factor of four whereas no difference in initial single shot soft-x-ray laser energy is observed.
A first explanation to this phenomenon would be some protective effect of the prepulse plasma that would prevent the target surface to be too strongly dammaged or drilled by laser ablation. The motion of the soft-x-ray laser source towards the bulk of the target was monitored with different series of near field images obtained for a same target position. For each near field image, the barycentre of the source spot is calculated after background and self emission subtraction. The position along the direction normal to the target drifts slowly towards the target interior. This evolution has been linearly fitted and an average penetration speed was estimated. This gives an estimation of the target surface ablation rate expressed in micrometer per shot. Ablation rate with respect to prepulse energy and compression are summarized in Fig. 5. This rate is greater in presence of a prepulse meaning that the target is drilled faster. However, above the 20 mJ threshold value, the ablation rate depends weakly on the prepulse energy. In conclusion, the prepulse does not help to protect the target against ablation.
Fig. 5 Evolution of the ablation rate (defined in the text) as a function of prepulse energy and prepulse compression.
The same near field images series enable us to measure the evolution of the vertical and horizontal source sizes (FWHM). They are relatively stable during the target lifetime (n < ntarget) and are presented in Fig. 6 as a function of EPP and for different values of τPP.
Fig. 6 Evolution of the soft-x-ray laser source size in the horizontal (left) and vertical (right) directions as a function of EPP and for different values of τPP.
The prepulse has little effect on the source width in the horizontal direction, which is independent of Epp and τpp However, the prepulse can lead to a significant increase in the vertical width (up to 60 percents). This is more pronounced with pulse compression of (82 ps) than with 32 ps or 188 ps.
For n significantly greater than ntarget, or when the soft-x-ray laser energy drops below 10 percent of its initial level E1 the source shape becomes unstable. The source can present randomly from shot to shot a double spot structure as depicted in Fig. 7b, or only one of these two spots. This unstable behavior occurs roughly when the soft-x-ray laser source has penetrated from one horizontal source width in the target bulk.
Fig. 7 Typical near field images of the soft-x-ray laser in the presence of a prepulse. Left: after 10 shots. Right: after 300 shots. The target bulk is on the left side. The horizontal and vertical scales are labeled in micrometers.
The mechanisms explaining the destruction of soft x-ray lasing could be the following: after each laser shot at the same place, the target surface becomes more and more concave. This can be checked by post-mortem target profilometry, or by extracting the target surface shape from near field images with contrast enhancement operations. Beyond a critical curvature radius, and therefore a critical ablation level, the plasma formation and expansion from such a concave surface may lead to electron density gradients unfavorable for soft-x-ray laser amplification. Two arguments are in favor of this interpretation. (i) During some series of shots, far field images of the soft-x-ray laser beam were recorded enabling to follow the soft-x-ray laser beam pointing. An increasing beam deflexion angle has been observed from the beginning to the end of the target irradiation, which might be associated with an increasing density gradient in the direction normal to the target plane. (ii) The formation of two soft-x-ray laser spots when the source is becoming weak might be related to some plasma flow collision occurring when the curvature radius is high. This collision can lead to the formation of a plasma jet in the middle of the soft-x-ray laser medium as observed by Purvis et al. [17].
As it is known that prepulses can reduce the plasma density gradients in the soft-x-ray laser gain region [22, 23], we postulate that this phenomenon might also allow the soft-x-ray laser to work with target surface presenting stronger concavity and thus might delay the end of soft-x-ray laser operation. Further analysis and numerical simulations are still needed to support this hypothesis. A similar experiment investigating the effect of line-focii widths may give additionnal informations on the target surface concavity effect.
5. Conclusion
This paper presented a study of prepulse effect on real 10 Hz operation of plasma based soft-x-ray laser. An promising regime as been found for an optimal prepulse delay of 1.4 ns and a prepulse energy above 10 mJ. In this regime, the soft-x-ray laser output energy has been increased by 50 % while the number of shots that can be performed at the same target location has been extended from 60 shots without prepulse to 300 shots in the optimal prepulse conditions. Moreover in this particular case, XUV energy is stable during the first 100 shots. The prepulse arrangement described in the present work enables a four time increase of the integrated XUV dose that can be produced by a single target position. A 200 μm translation of the target enables to restart a cycle in the optimal conditions, so that a single 50 mm mechanical polishing grade Mo target slab enables 75000 shots or roughly two hours of operation (not taking into account the target motion time between cycles) in optimal conditions.
Since the optimization parameters have been found, the prepulse generation could be greatly simplified by generating it at the beginning of the laser chain in the same way as pumping pulses [14]. However, three issues would need to be investigated before a routine operation. The first one is related to the influence of the prepulse angle of incidence on target [16]. The second is related to the beam pointing evolution as target is drilled. Finally, as mentioned in the introduction, plasma based soft-x-ray laser offers a variety of wavelength by changing the target material. Transposition of these results to other material is not straighforward as solid material properties, namely melting point and heat conductivity can dramatically affect the preplasma generation [18].
Acknowledgments
Jamil Habib acknowledges the financial support of the Extreme Light Infrastructure (ELI). Multilayer optics for soft-x-ray laser manipulations were provided by the CEMOX facility. The support of the Agence Nationale de la Recherche (ANR) is also acknowledge through the project “jeunes chercheuses et jeunes chercheurs” ASOURIX ANR-09-JCJC-0056.
References and links
1. Y. Wang, M. A. Larotonda, B. M. Luther, D. Alessi, M. Berrill, V. N. Shlyaptsev, and J. J. Rocca, “Demonstration of high-repetition-rate tabletop soft-x-ray lasers with saturated output at wavelengths down to 13.9 nm and gain down to 10.9 nm,” Phys. Rev. A 2, 053807 (2005). [CrossRef]
2. R. F. Smith, J. Dunn, J. Nilsen, V. N. Shlyaptsev, S. Moon, J. Filevich, J. J. Rocca, M. C. Marconi, J. R. Hunter, and T. W. Barbee Jr, “Picosecond soft-x-ray laser interferometry of dense plasmas,” Phys. Rev. E 89, 065004 (2002).
3. M. H. Edwards, D. S. Whittaker, G. J. Tallents, P. Mistry, G. J. Pert, B. Rus, T. Mocek, M. Kozlova, J. Polan, A. Praeg, M. Stupka, and P. Homer, “Laser-ablation rates measured using soft-x-ray laser transmission,” Phys. Rev. Lett. 99, 195002 (2007).
4. M. G. Capeluto, G. Vaschenko, M. Grisham, M. C. Marconi, S. Luduena, L. Pietrasanta, Y. Lu, B. Parkinson, C. S. Menoni, and J. J. Rocca, “Nanopatterning with interferometric lithography using a compact λ = 46.9nm Laser,” IEEE Trans. Nanotechnol. 5, 3–7 (2006). [CrossRef]
5. B. Zielbauer, D. Zimmer, J. Habib, O. Guilbaud, S. Kazamias, M. Pittman, and D. Ros, “Stable and fully controlled long-time operation of a soft X-ray laser for user application experiments,” Appl. Phys. B 100, 731–736 (2010). [CrossRef]
6. F. Tissandier, S. Sebban, M. Ribiere, J. Gautier, Ph. Zeitoun, G. Lambert, A. Barszczak Sardinha, J.-Ph. Goddet, F. Burgy, T. Lefrou, C. Valentin, A. Rousse, O. Guilbaud, A. Klisnick, J. Nejdl, T. Mocek, and G. Maynard, “Observation of spectral gain narrowing in a high-order harmonic seeded soft-x-ray amplifier,” Phys. Rev. A 81, 063833 (2010). [CrossRef]
7. L. M. R. Gartside, G. J. Tallents, A. K. Rossall, E. Wagenaars, D. S. Whittaker, M. Kozlova, J. Nejdl, M. Sawicka, J. Polan, M. Kalal, and B. Rus, “Extreme ultraviolet interferometry of warm dense matter in laser plasmas,” Opt. Lett. 35, 3820–3822 (2010). [CrossRef] [PubMed]
8. J. J. Rocca, O. D. Cortazar, B. Szapiro, K. Floyd, and F. G. Tomasel, “Fast-discharge excitation of hot capillary plasmas for soft-x-ray amplifiers,” Phys. Rev. E 47, 1299–1304 (1993). [CrossRef]
9. S. Sebban, R. Haroutunian, P. Balcou, G. Grillon, A. Rousse, S. Kazamias, T. Marin, J. P. Rousseau, L. Notebaert, M. Pittman, J. P. Chambaret, A. Antonetti, D. Hulin, D. Ros, A. Klisnick, A. Carillon, P. Jaegle, G. Jamelot, and J.F. Wyart, “Saturated amplification of a collisionally pumped optical-field-ionization soft X-Ray laser at 41.8 nm,” Phys. Rev. Lett. 86, 3004–3007 (2001). [CrossRef] [PubMed]
10. D. Alessi, D. H. Martz, Y. Wang, M. Berrill, B. M. Luther, and J. J. Rocca, “Gain-saturated 10.9 nm tabletop laser operating at 1 Hz repetition rate,” Opt. Lett. 35414–416 (2010). [CrossRef] [PubMed]
11. T. Kawachi, A. Sasaki, M. Tanaka, M. Kishimoto, N. Hasegawa, K. Nagashima, M. Koike, H. Daido, and Y. Kato, “Observation of strong soft-x-ray amplification at 8.8 nm in the transient collisional-excitation scheme,” Phys Rev. A 69, 033805 (2004). [CrossRef]
12. A. Weith, M. A. Larotonda, Y. Wang, B. M. Luther, D. Alessi, M. C. Marconi, J. J. Rocca, and J. Dunn, “Continuous high-repetition-rate operation of collisional soft-x-ray lasers with solid targets,” Opt. Lett. 31, 1994–1996 (2006). [CrossRef] [PubMed]
13. H. Stiel, J. Tmmler, R. Jung, P. V. Nickles, and W. Sandner, “X-ray laser takes the 100 Hz barrier,” in Soft X-Ray Lasers and Applications VIII, J. Dunn and G.J. Tallents, eds., Proc. SPIE7451, 745109 (2009).
14. D. Zimmer, B. Zielbauer, M. Pittman, O. Guilbaud, J. Habib, S. Kazamias, D. Ros, V. Bagnoud, and T. Kuehl, “Optimization of a tabletop high-repetition-rate soft-x-ray laser pumped in double-pulse single-beam grazing incidence,” Opt. Lett. 35, 450–452 (2010). [CrossRef] [PubMed]
15. R. Keenan, J. Dunn, P. K. Patel, D. F. Price, R. F. Smith, and V. N. Shlyaptsev, “High-repetition-rate grazing-incidence pumped soft-x-ray laser operating at 18.9 nm,” Phys. Rev. Lett. 94, 103901 (2005). [CrossRef] [PubMed]
16. K. A. Janulewicz and C. M. Kim, “Role of the precursor in a triple-pulse pumping scheme of a nickel-like silver soft-x-ray laser in the grazing-incidence-pumping geometry,” Phys. Rev. E 82056405 (2010). [CrossRef]
17. M. Purvis, J. Grava, J. Filevich, M. C. Marconi, J. Dunn, S. J. Moon, V. N. Shlyaptsev, E. Jankowska, and J. J. Rocca, “Dynamics of converging laser-created plasmas in semicylindrical cavities studied using soft-x-ray laser interferometry,” Phys. Rev. E 76046402 (2007). [CrossRef]
18. B. Rus, P. Zeitoun, T. Mocek, S. Sebban, M. Kalal, A. Demir, G. Jamelot, A. Klisnick, B. Kralikova, J. Skala, and G. J. Tallents, “Investigation of Zn and Cu prepulse plasmas relevant to collisional excitation soft x-ray lasers,” Phys. Rev. A 564229–4241(1997). [CrossRef]
19. M. A. Larotonda, B. M. Luther, Y. Wang, Y. Liu, D. Alessi, M. Berrill, A. Dummer, F. Brizuela, C. S. Menoni, M. C. Marconi, V. N. Shlyaptsev, J. Dunn, and J. J. Rocca, “Characteristics of a saturated 18.9-nm tabletop laser operating at 5-Hz repetition rate,” IEEE J. of sel. top. in quant. elect. 10, 1363–1367 (2004). [CrossRef]
20. K. Cassou, S. Kazamias, D. Ros, F. Pl, G. Jamelot, A. Klisnick, O. Lundh, F. Lindau, A. Persson, C.-G. Wahlstrm, S. de Rossi, D. Joyeux, B. Zielbauer, D. Ursescu, and T. Khl, “Optimization toward a high-average brightness soft-x-ray laser pumped at grazing incidence,” Opt. Lett. 32, 139–141 (2007). [CrossRef]
21. D. Alessi, B. M. Luther, Y. Wang, M. A. Larotonda, M. Berrill, and J. J. Rocca, “High repetition rate operation of saturated tabletop soft x-ray lasers in transitions of neon-like ions near 30 nm,” Opt. Express 13, 2093–2098, (2005). [CrossRef] [PubMed]
22. J. Nilsen, B. J. MacGowan, L. B. Da Silva, and J. C. Moreno, “Prepulse technique for producing low-Z Ne-like x-ray lasers,” Phys. Rev. A 48, 4682–4685 (1993). [CrossRef] [PubMed]
23. R. Tommasini, K. Eidmann, T. Kawachi, and E. Fill, “Preplasma conditions for operation of 10-Hz subjoule femtosecond-laser-pumped nickel-like soft x-ray lasers,” Phys. Rev. E 69, 066404 (2007). [CrossRef]
