Note: Enhancement of the extreme ultraviolet emission from a potassium plasma by dual laser irradiation Takeshi Higashiguchi, Mami Yamaguchi, Takamitsu Otsuka, Takeshi Nagata, Hayato Ohashi, Bowen Li, Rebekah D’Arcy, Padraig Dunne, and Gerry O’Sullivan Citation: Review of Scientific Instruments 85, 096102 (2014); doi: 10.1063/1.4894384 View online: http://dx.doi.org/10.1063/1.4894384 View Table of Contents: http://scitation.aip.org/content/aip/journal/rsi/85/9?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Three-dimensional extreme ultraviolet emission from a droplet-based laser-produced plasma J. Appl. Phys. 114, 033303 (2013); 10.1063/1.4815955 13.5 nm extreme ultraviolet emission from tin based laser produced plasma sources J. Appl. Phys. 99, 093302 (2006); 10.1063/1.2191477 Enhancement of extreme ultraviolet emission from a lithium plasma by use of dual laser pulses Appl. Phys. Lett. 88, 161502 (2006); 10.1063/1.2195904 Enhancement of laser plasma extreme ultraviolet emission by shockwave-laser interaction Phys. Plasmas 12, 042701 (2005); 10.1063/1.1857914 Characterization of extreme ultraviolet emission from laser-produced spherical tin plasma generated with multiple laser beams Appl. Phys. Lett. 86, 051501 (2005); 10.1063/1.1856697
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 141.212.109.170 On: Fri, 21 Nov 2014 18:14:04
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 096102 (2014)
Note: Enhancement of the extreme ultraviolet emission from a potassium plasma by dual laser irradiation Takeshi Higashiguchi,1,a) Mami Yamaguchi,1 Takamitsu Otsuka,1 Takeshi Nagata,1 Hayato Ohashi,2 Bowen Li,3,4 Rebekah D’Arcy,4 Padraig Dunne,4 and Gerry O’Sullivan4 1 Department of Advanced Interdisciplinary Sciences and Center for Optical Research (CORE), Utsunomiya University, Yoto 7-1-2, Utsunomiya, Tochigi 321-8585 Japan 2 Graduate School of Science and Engineering for Research, University of Toyama, Toyama, Toyama 930-8555, Japan 3 School of Nuclear Science and Technology, Lanzhou University, Lanzhou, 730000, China 4 School of Physics, University College Dublin, Belfield, Dublin 4, Ireland
(Received 25 July 2014; accepted 19 August 2014; published online 3 September 2014) Emission spectra from multiply charged potassium ions ranging from K3+ to K5+ have been obtained in the extreme ultraviolet (EUV) spectral region. A strong emission feature peaking around 38 nm, corresponding to a photon energy of 32.6 eV, is the dominant spectral feature at time-averaged electron temperatures in the range of 8−12 eV. The variation of this emission with laser intensity and the effects of pre-pulses on the relative conversion efficiency (CE) have been explored experimentally and indicate that an enhancement of about 30% in EUV CE is readily attainable. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4894384] Considerable effort is currently being expended on the development of highly charged ion plasma sources for the generation of extreme ultraviolet (EUV) and soft x-ray radiation for applications such as photo-absorption spectroscopy, semiconductor lithography, and biological imaging.1–10 Photon energies in excess of 10 eV are sufficient to induce photochemical interactions in almost all organic molecules or on solid surfaces.11, 12 Also, since SiO2 has strong absorption coefficient around 40 nm, it is expected that micromachining by EUV illumination at a wavelength around 40 nm is possible.13 Plasmas containing potassium are known to emit in this wavelength region.14 Thus, for surface morphology applications that rely on photochemical reactions, it is important to understand the EUV emission processes in potassium plasmas and their dependence on plasma conditions in order to optimize their emission energy. The observation of intense EUV emission centered around 38 nm with a bandwidth of 8 nm [full width at halfmaximum (FWHM)] produced from a hollow-cathode microdischarge containing potassium has been demonstrated.14 By comparison with atomic structure calculations and previously reported spectra,15, 16 the broadband emission was found to be primarily due to 3p−3d transitions in potassium ions ranging from K2+ to K5+ . The spectrum of this capillary discharge-produced potassium plasma suffered from some contamination that gave rise to strong line emission around 38 nm from carbon, due to ablation of the capillary wall material [polytetrafluoroethylene (PTFE)]. Therefore to characterize the spectral component due to potassium alone, we subsequently compared spectra from discharge-produced plasmas with those of laser-produced plasmas.17 The spectra recorded at similar electron temperatures were found to be almost identical. The temporal behavior of the emission from a laser-produced plasma was also studied and compared a) [email protected]
0034-6748/2014/85(9)/096102/3/$30.00
with predictions from a numerical simulation based on a timedependent collisional radiative (CR) model.18 The dominant emission was shown to be produced when the time-averaged electron temperature was about 10 eV. The results further showed that the highest ion stages attained initially were K5+ to K7+ ; thereafter the plasma recombined to lower ionization states that emitted in the 38 nm band. Reheating of such recombining plasmas by a second laser pulse has been shown to lead to a significant increase in overall emission efficiency (CE) defined as the fraction of laser energy converted to inband EUV energy.6, 9 As the dominant emission energy is radiated during the recombination phase at lower electron temperatures, the plasma parameters should be tuned by controlling the laser intensity and delay time (pulse separation time) in a dual laser pulse irradiation scheme, for optimization of the excited ionic charge state populations for 38-nm emission. With this in mind, it is important to record the spectral variation for multiply charge state potassium plasmas obtained for different irradiation conditions. In this report, we describe the results of a spectral analysis undertaken in order to evaluate the optimum electron temperature. We also explore the dependence of EUV CE on intensity and on the efficacy of dual pulse irradiation for the improvement of the CE of the 38-nm EUV source. A Q-switched Nd:YAG (Nd:yttrium-aluminum-garnet) laser operating at 1064 nm with a maximum pulse energy of 2 J and a duration of 10 ns (FWHM) was focused, using a 12-cm focal length lens, onto a planar potassium target with a thickness of 1 mm inside a vacuum chamber. The spot size was measured to be 500 μm (FWHM). This spot size was chosen to minimize the plasma hydrodynamic expansion loss.19 The pulse energy at the fixed spot diameter of 500 μm was adjusted to vary the laser intensity. The emission spectrum was recorded with a normal incidence vacuum spectrometer using an iridium coated grating with 1200 lines/mm (Acton: VM502). Time-averaged spectra were obtained
85, 096102-1
© 2014 AIP Publishing LLC
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 141.212.109.170 On: Fri, 21 Nov 2014 18:14:04
096102-2
Higashiguchi et al.
Rev. Sci. Instrum. 85, 096102 (2014)
FIG. 1. Comparison of the plasma emission spectrum with experimentally measured line strength data weighted by ion stage assuming steady state collisional radiative equilibrium (CR) for plasma electron temperatures of 6 (a), 7 (b), 8 (c), 10 (d), 12 (e), and 15 eV (f), respectively.
again for power densities in excess of 8 × 1010 W/cm2 . Theoretical simulation, based on a numerical evaluation that couples the CR model with atomic structure calculations for line emission and rate coefficients,20–22 predicts that the CE should peak close to a power density of 4.5 × 1010 W/cm2 . At these laser intensities the electron temperature lies in the 10−15 eV range, and the corresponding ion populations may be inferred as being dominated by K4+ and K5+ which have 3p3 and 3p2 ground state configurations.18 These stages emit strongly in the 38 nm region. Any increase in laser power shifts the 2.0
CE (arb. units)
using a thermoelectrically cooled back-illuminated x-ray CCD (charge coupled device) camera. The typical spectral resolution was better than 0.5 nm. The pressure in the spectrometer was also maintained at less than 3 × 10−5 Pa. The emission from the plasma in the EUV spectral region was viewed at 45◦ with respect to the incident laser axis. The spectrum obtained at a focussed pulse power density 2 × 1010 W/cm2 is presented in Fig. 1. The ion stage distribution was calculated using a time-dependent CR model for an electron density of 1 × 1020 cm−3 in a plasma that was assumed to be optically thin to EUV radiation. The experimentally determined gA values (g and A are, respectively, the statistical weight and the Einstein spontaneous emission coefficient) for the strongest emission lines15, 16 weighted by ion stage for a plasma characterised by average electron temperatures in the range between 6 and 15 eV are presented in Fig. 1. Since the plasma emission spectrum is time averaged over the entire plasma lifetime, it is seen from this figure that an average value in the 8−12 eV range best characterises the bulk of the emission spectrum. The variation of the relative energy CE from the laser energy to the EUV emission energy at 38 nm within a bandwidth of ±4 nm for single laser pulse irradiation is shown in Fig. 2. The maximum CE is attained at a flux of 3 × 1010 W/cm2 and thereafter decreases though with a possible slight increase
1.5
1.0
0.5
0
0
2
4
6
8
10
12
Laser intensity (x1010 W/cm2 ) FIG. 2. Measured and calculated conversion efficiencies of laser to EUV energy within an 8 nm bandwidth at 38 nm.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 141.212.109.170 On: Fri, 21 Nov 2014 18:14:04
096102-3
Higashiguchi et al.
Rev. Sci. Instrum. 85, 096102 (2014)
Electron temp. (eV)
XUV CE (arb. units)
Dual laser pulses Single pulse Electron temperature
Pulse separation time (ns) FIG. 3. Calculated variation of the electron temperature and measured conversion efficiency of a potassium dual-pulse laser produced plasma as a function of the temporal separation of the pulses.
average degree of ionization to higher values. Since resonant 3p − 3d transitions are responsible for the observed emission, the opacity of these increases as the plasma temperature increases. In addition, the 3p shell is fully stripped in K7+ so ions past K6+ will not contribute significantly to the spectra. In order to enhance the EUV CE for production of an efficient EUV source, a dual laser-produced plasma experiment was performed to study the effects of pulse separation time dependence on the 38-nm in-band emission. In earlier work on Nd:YAG laser-produced Li and SnO2 plasmas for 13.5 nm operation and Gd plasmas for 6.7 nm operation, the dual laser pulse irradiation technique was used to enhance the emission intensity, by providing targets with lower initial density and opacity.6, 23, 24 The use of dual laser pulses is an efficient way to control plasma dynamics. The scheme would allow the effective utilization of the main laser pulse energy to heat a pre-plasma, which would otherwise be wasted on ionizing the target material. Choice of an appropriate delay between dual laser pulses could in effect control the electron density of a plasma and the scale length of its density gradient, where the main laser pulse would interact. Two laser pulses of 8 and 10 ns (FWHM) at 532 and 1064 nm, were provided as the pre-pulse (532 nm) and the main (1064 nm) laser pulse, respectively. The focused intensity of the main laser pulse was fixed at 2 × 1010 W/cm2 . Figure 3 shows the observed relative CEs with the calculated electron temperatures as a function of the temporal separation of the pulses. From the numerical calculation the optimum pulse separation time required to produce approximately a 30% increase in electron temperature was predicted to be about 70 ns. In practice, an enhancement of close to 30% in EUV CE was observed after approximately 20 ns, while a figure close to 30% was again obtained after a delay of approximately 200 ns. Thus it is difficult to correlate the enhanced EUV CE directly with increased electron temperature. As with the results of previous studies, the primary source of enhancement results from interaction with a plasma volume still sufficiently dense to guarantee laser plasma interaction, while the increased plasma energy at short interpulse delays plays a subsidiary role. It can be inferred that for dual pulses for interpulse delays ranging from 0 to 250 ns a gain of about 30% in overall EUV CE is obtained.
In summary, we have observed an enhancement of the EUV emission at 38 nm from laser-produced potassium plasmas and compared experimental trends with the results of detailed calculations. We compared the predictions for electron temperature and the variation of EUV CE with laser pulse power density and for the effects of using dual pulse irradiation on the overall EUV CE with experimentally determined values. An enhancement of about 30% in EUV CE was observed after approximately 20 ns. In the near future, we plan to develop a compact potassium plasma EUV source, coupled with a Sc/Si multilayer mirror for use in a photoinduced desorption spectrometer.25, 26 A part of this work was performed under “Project for Bio-Imaging and Sensing at Utsunomiya University” from MEXT. The UCD group was supported by Science Foundation Ireland International Co-operation Strategic Award 13/ISCA/2846. 1 R.
D’Arcy, J. T. Costello, E. T. Kennedy, C. McGuinness, J. P. Mosnier, and G. O’Sullivan, J. Phys. B 33, 1383 (2000). 2 EUV Sources for Lithography, edited by V. Bakshi (SPIE, Bellingham, WA, 2005). 3 P. A. C. Jansson, U. Vogt, and H. M. Hertz, Rev. Sci. Instrum. 76, 043503 (2005). 4 P. A. C. Takman, H. Stollberg, G. A. Johansson, A. Holmberg, M. Lindblom, and H. M. Hertz, J. Microsc. 226, 175 (2007). 5 T. Otsuka, D. Kilbane, J. White, T. Higashiguchi, N. Yugami, T. Yatagai, W. Jiang, A. Endo, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 97, 111503 (2010). 6 T. Higashiguchi, T. Otsuka, N. Yugami, T. Yatagai, W. Jiang, A. Endo, D. Kilbane, B. W. Li, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 99, 191502 (2011). 7 G. Tallents, E. Wagenaars, and G. Pert, Nat. Photonics 4, 809 (2010). 8 T. Cummins, T. Otsuka, N. Yugami, W. Jiang, A. Endo, B. W. Li, C. O’Gorman, P. Dunne, E. Sokell, G. O’Sullivan, and T. Higashiguchi, Appl. Phys. Lett. 100, 061118 (2012). 9 G. O’Sullivan, D. Kilbane, and R. D’Arcy, J. Mod. Opt. 59, 855 (2012). 10 T. Higashiguchi, T. Otsuka, N. Yugami, W. Jiang, A. Endo, B. W. Li, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 100, 014103 (2012). 11 T. Ohtsubo, T. Azuma, M. Takaura, T. Higashiguchi, S. Kubodera, and W. Sasaki, Appl. Phys. A 76, 139 (2003). 12 Y. Maezono, K. Toshikawa, K. Kurosawa, K. Amari, S. Ishimura, M. Katto, and A. Yokotani, Jpn. J. Appl. Phys. 46, 3534 (2007). 13 H. Akazawa, J. Takahashi, Y. Utsumi, I. Kawashima, and T. Urisu, J. Vac. Sci. Technol. A 9, 2653 (1991). 14 T. Higashiguchi, H. Terauchi, N. Yugami, T. Yatagai, W. Sasaki, R. D’Arcy, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 96, 131505 (2010). 15 J. Sugar and C. Corliss, J. Phys. Chem. Ref. Data 14(Suppl. 2), 1 (1985). 16 J. E. Sansonetti, J. Phys. Chem. Ref. Data 37, 7 (2008). 17 T. Higashiguchi, H. Terauchi, T. Otsuka, M. Yamaguchi, K. Kikuchi, N. Yugami, T. Yatagai, W. Sasaki, R. D’Arcy, P. Dunne, and G. O’Sullivan, J. Appl. Phys. 109, 013301 (2011). 18 T. Higashiguchi, M. Yamaguchi, T. Otsuka, H. Terauchi, N. Yugami, T. Yatagai, R. D’Arcy, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 98, 091503 (2011). 19 R. C. Spitzer, T. J. Orzechowski, D. W. Phillion, R. L. Kauffman, and C. Cerjan, J. Appl. Phys. 79, 2251 (1996). 20 D. Colombant and G. F. Tonon, J. Appl. Phys. 44, 3524 (1973). 21 T. Fujimoto, J. Phys. Soc. Jpn. 47, 265 (1979); 47, 273 (1979). 22 H. Griem, Principle of Plasma Spectroscopy (Cambridge Monographs on Plasma Physics, 2005). 23 T. Higashiguchi, K. Kawasaki, W. Sasaki, and S. Kubodera, Appl. Phys. Lett. 88, 161502 (2006). 24 T. Higashiguchi, N. Dojyo, M. Hamada, W. Sasaki, and S. Kubodera, Appl. Phys. Lett. 88, 201503 (2006). 25 N. Hirashita, M. Kinoshita, I. Aikawa, and T. Ajioka, Appl. Phys. Lett. 56, 451 (1990). 26 M. Wasamoto, M. Katto, M. Kaku, S. Kubodera, and A. Yokotani, Appl. Surf. Sci. 255, 9861 (2009).
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 141.212.109.170 On: Fri, 21 Nov 2014 18:14:04
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 141.212.109.170 On: Fri, 21 Nov 2014 18:14:04
REVIEW OF SCIENTIFIC INSTRUMENTS 85, 096102 (2014)
Note: Enhancement of the extreme ultraviolet emission from a potassium plasma by dual laser irradiation Takeshi Higashiguchi,1,a) Mami Yamaguchi,1 Takamitsu Otsuka,1 Takeshi Nagata,1 Hayato Ohashi,2 Bowen Li,3,4 Rebekah D’Arcy,4 Padraig Dunne,4 and Gerry O’Sullivan4 1 Department of Advanced Interdisciplinary Sciences and Center for Optical Research (CORE), Utsunomiya University, Yoto 7-1-2, Utsunomiya, Tochigi 321-8585 Japan 2 Graduate School of Science and Engineering for Research, University of Toyama, Toyama, Toyama 930-8555, Japan 3 School of Nuclear Science and Technology, Lanzhou University, Lanzhou, 730000, China 4 School of Physics, University College Dublin, Belfield, Dublin 4, Ireland
(Received 25 July 2014; accepted 19 August 2014; published online 3 September 2014) Emission spectra from multiply charged potassium ions ranging from K3+ to K5+ have been obtained in the extreme ultraviolet (EUV) spectral region. A strong emission feature peaking around 38 nm, corresponding to a photon energy of 32.6 eV, is the dominant spectral feature at time-averaged electron temperatures in the range of 8−12 eV. The variation of this emission with laser intensity and the effects of pre-pulses on the relative conversion efficiency (CE) have been explored experimentally and indicate that an enhancement of about 30% in EUV CE is readily attainable. © 2014 AIP Publishing LLC. [http://dx.doi.org/10.1063/1.4894384] Considerable effort is currently being expended on the development of highly charged ion plasma sources for the generation of extreme ultraviolet (EUV) and soft x-ray radiation for applications such as photo-absorption spectroscopy, semiconductor lithography, and biological imaging.1–10 Photon energies in excess of 10 eV are sufficient to induce photochemical interactions in almost all organic molecules or on solid surfaces.11, 12 Also, since SiO2 has strong absorption coefficient around 40 nm, it is expected that micromachining by EUV illumination at a wavelength around 40 nm is possible.13 Plasmas containing potassium are known to emit in this wavelength region.14 Thus, for surface morphology applications that rely on photochemical reactions, it is important to understand the EUV emission processes in potassium plasmas and their dependence on plasma conditions in order to optimize their emission energy. The observation of intense EUV emission centered around 38 nm with a bandwidth of 8 nm [full width at halfmaximum (FWHM)] produced from a hollow-cathode microdischarge containing potassium has been demonstrated.14 By comparison with atomic structure calculations and previously reported spectra,15, 16 the broadband emission was found to be primarily due to 3p−3d transitions in potassium ions ranging from K2+ to K5+ . The spectrum of this capillary discharge-produced potassium plasma suffered from some contamination that gave rise to strong line emission around 38 nm from carbon, due to ablation of the capillary wall material [polytetrafluoroethylene (PTFE)]. Therefore to characterize the spectral component due to potassium alone, we subsequently compared spectra from discharge-produced plasmas with those of laser-produced plasmas.17 The spectra recorded at similar electron temperatures were found to be almost identical. The temporal behavior of the emission from a laser-produced plasma was also studied and compared a) [email protected]
0034-6748/2014/85(9)/096102/3/$30.00
with predictions from a numerical simulation based on a timedependent collisional radiative (CR) model.18 The dominant emission was shown to be produced when the time-averaged electron temperature was about 10 eV. The results further showed that the highest ion stages attained initially were K5+ to K7+ ; thereafter the plasma recombined to lower ionization states that emitted in the 38 nm band. Reheating of such recombining plasmas by a second laser pulse has been shown to lead to a significant increase in overall emission efficiency (CE) defined as the fraction of laser energy converted to inband EUV energy.6, 9 As the dominant emission energy is radiated during the recombination phase at lower electron temperatures, the plasma parameters should be tuned by controlling the laser intensity and delay time (pulse separation time) in a dual laser pulse irradiation scheme, for optimization of the excited ionic charge state populations for 38-nm emission. With this in mind, it is important to record the spectral variation for multiply charge state potassium plasmas obtained for different irradiation conditions. In this report, we describe the results of a spectral analysis undertaken in order to evaluate the optimum electron temperature. We also explore the dependence of EUV CE on intensity and on the efficacy of dual pulse irradiation for the improvement of the CE of the 38-nm EUV source. A Q-switched Nd:YAG (Nd:yttrium-aluminum-garnet) laser operating at 1064 nm with a maximum pulse energy of 2 J and a duration of 10 ns (FWHM) was focused, using a 12-cm focal length lens, onto a planar potassium target with a thickness of 1 mm inside a vacuum chamber. The spot size was measured to be 500 μm (FWHM). This spot size was chosen to minimize the plasma hydrodynamic expansion loss.19 The pulse energy at the fixed spot diameter of 500 μm was adjusted to vary the laser intensity. The emission spectrum was recorded with a normal incidence vacuum spectrometer using an iridium coated grating with 1200 lines/mm (Acton: VM502). Time-averaged spectra were obtained
85, 096102-1
© 2014 AIP Publishing LLC
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 141.212.109.170 On: Fri, 21 Nov 2014 18:14:04
096102-2
Higashiguchi et al.
Rev. Sci. Instrum. 85, 096102 (2014)
FIG. 1. Comparison of the plasma emission spectrum with experimentally measured line strength data weighted by ion stage assuming steady state collisional radiative equilibrium (CR) for plasma electron temperatures of 6 (a), 7 (b), 8 (c), 10 (d), 12 (e), and 15 eV (f), respectively.
again for power densities in excess of 8 × 1010 W/cm2 . Theoretical simulation, based on a numerical evaluation that couples the CR model with atomic structure calculations for line emission and rate coefficients,20–22 predicts that the CE should peak close to a power density of 4.5 × 1010 W/cm2 . At these laser intensities the electron temperature lies in the 10−15 eV range, and the corresponding ion populations may be inferred as being dominated by K4+ and K5+ which have 3p3 and 3p2 ground state configurations.18 These stages emit strongly in the 38 nm region. Any increase in laser power shifts the 2.0
CE (arb. units)
using a thermoelectrically cooled back-illuminated x-ray CCD (charge coupled device) camera. The typical spectral resolution was better than 0.5 nm. The pressure in the spectrometer was also maintained at less than 3 × 10−5 Pa. The emission from the plasma in the EUV spectral region was viewed at 45◦ with respect to the incident laser axis. The spectrum obtained at a focussed pulse power density 2 × 1010 W/cm2 is presented in Fig. 1. The ion stage distribution was calculated using a time-dependent CR model for an electron density of 1 × 1020 cm−3 in a plasma that was assumed to be optically thin to EUV radiation. The experimentally determined gA values (g and A are, respectively, the statistical weight and the Einstein spontaneous emission coefficient) for the strongest emission lines15, 16 weighted by ion stage for a plasma characterised by average electron temperatures in the range between 6 and 15 eV are presented in Fig. 1. Since the plasma emission spectrum is time averaged over the entire plasma lifetime, it is seen from this figure that an average value in the 8−12 eV range best characterises the bulk of the emission spectrum. The variation of the relative energy CE from the laser energy to the EUV emission energy at 38 nm within a bandwidth of ±4 nm for single laser pulse irradiation is shown in Fig. 2. The maximum CE is attained at a flux of 3 × 1010 W/cm2 and thereafter decreases though with a possible slight increase
1.5
1.0
0.5
0
0
2
4
6
8
10
12
Laser intensity (x1010 W/cm2 ) FIG. 2. Measured and calculated conversion efficiencies of laser to EUV energy within an 8 nm bandwidth at 38 nm.
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 141.212.109.170 On: Fri, 21 Nov 2014 18:14:04
096102-3
Higashiguchi et al.
Rev. Sci. Instrum. 85, 096102 (2014)
Electron temp. (eV)
XUV CE (arb. units)
Dual laser pulses Single pulse Electron temperature
Pulse separation time (ns) FIG. 3. Calculated variation of the electron temperature and measured conversion efficiency of a potassium dual-pulse laser produced plasma as a function of the temporal separation of the pulses.
average degree of ionization to higher values. Since resonant 3p − 3d transitions are responsible for the observed emission, the opacity of these increases as the plasma temperature increases. In addition, the 3p shell is fully stripped in K7+ so ions past K6+ will not contribute significantly to the spectra. In order to enhance the EUV CE for production of an efficient EUV source, a dual laser-produced plasma experiment was performed to study the effects of pulse separation time dependence on the 38-nm in-band emission. In earlier work on Nd:YAG laser-produced Li and SnO2 plasmas for 13.5 nm operation and Gd plasmas for 6.7 nm operation, the dual laser pulse irradiation technique was used to enhance the emission intensity, by providing targets with lower initial density and opacity.6, 23, 24 The use of dual laser pulses is an efficient way to control plasma dynamics. The scheme would allow the effective utilization of the main laser pulse energy to heat a pre-plasma, which would otherwise be wasted on ionizing the target material. Choice of an appropriate delay between dual laser pulses could in effect control the electron density of a plasma and the scale length of its density gradient, where the main laser pulse would interact. Two laser pulses of 8 and 10 ns (FWHM) at 532 and 1064 nm, were provided as the pre-pulse (532 nm) and the main (1064 nm) laser pulse, respectively. The focused intensity of the main laser pulse was fixed at 2 × 1010 W/cm2 . Figure 3 shows the observed relative CEs with the calculated electron temperatures as a function of the temporal separation of the pulses. From the numerical calculation the optimum pulse separation time required to produce approximately a 30% increase in electron temperature was predicted to be about 70 ns. In practice, an enhancement of close to 30% in EUV CE was observed after approximately 20 ns, while a figure close to 30% was again obtained after a delay of approximately 200 ns. Thus it is difficult to correlate the enhanced EUV CE directly with increased electron temperature. As with the results of previous studies, the primary source of enhancement results from interaction with a plasma volume still sufficiently dense to guarantee laser plasma interaction, while the increased plasma energy at short interpulse delays plays a subsidiary role. It can be inferred that for dual pulses for interpulse delays ranging from 0 to 250 ns a gain of about 30% in overall EUV CE is obtained.
In summary, we have observed an enhancement of the EUV emission at 38 nm from laser-produced potassium plasmas and compared experimental trends with the results of detailed calculations. We compared the predictions for electron temperature and the variation of EUV CE with laser pulse power density and for the effects of using dual pulse irradiation on the overall EUV CE with experimentally determined values. An enhancement of about 30% in EUV CE was observed after approximately 20 ns. In the near future, we plan to develop a compact potassium plasma EUV source, coupled with a Sc/Si multilayer mirror for use in a photoinduced desorption spectrometer.25, 26 A part of this work was performed under “Project for Bio-Imaging and Sensing at Utsunomiya University” from MEXT. The UCD group was supported by Science Foundation Ireland International Co-operation Strategic Award 13/ISCA/2846. 1 R.
D’Arcy, J. T. Costello, E. T. Kennedy, C. McGuinness, J. P. Mosnier, and G. O’Sullivan, J. Phys. B 33, 1383 (2000). 2 EUV Sources for Lithography, edited by V. Bakshi (SPIE, Bellingham, WA, 2005). 3 P. A. C. Jansson, U. Vogt, and H. M. Hertz, Rev. Sci. Instrum. 76, 043503 (2005). 4 P. A. C. Takman, H. Stollberg, G. A. Johansson, A. Holmberg, M. Lindblom, and H. M. Hertz, J. Microsc. 226, 175 (2007). 5 T. Otsuka, D. Kilbane, J. White, T. Higashiguchi, N. Yugami, T. Yatagai, W. Jiang, A. Endo, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 97, 111503 (2010). 6 T. Higashiguchi, T. Otsuka, N. Yugami, T. Yatagai, W. Jiang, A. Endo, D. Kilbane, B. W. Li, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 99, 191502 (2011). 7 G. Tallents, E. Wagenaars, and G. Pert, Nat. Photonics 4, 809 (2010). 8 T. Cummins, T. Otsuka, N. Yugami, W. Jiang, A. Endo, B. W. Li, C. O’Gorman, P. Dunne, E. Sokell, G. O’Sullivan, and T. Higashiguchi, Appl. Phys. Lett. 100, 061118 (2012). 9 G. O’Sullivan, D. Kilbane, and R. D’Arcy, J. Mod. Opt. 59, 855 (2012). 10 T. Higashiguchi, T. Otsuka, N. Yugami, W. Jiang, A. Endo, B. W. Li, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 100, 014103 (2012). 11 T. Ohtsubo, T. Azuma, M. Takaura, T. Higashiguchi, S. Kubodera, and W. Sasaki, Appl. Phys. A 76, 139 (2003). 12 Y. Maezono, K. Toshikawa, K. Kurosawa, K. Amari, S. Ishimura, M. Katto, and A. Yokotani, Jpn. J. Appl. Phys. 46, 3534 (2007). 13 H. Akazawa, J. Takahashi, Y. Utsumi, I. Kawashima, and T. Urisu, J. Vac. Sci. Technol. A 9, 2653 (1991). 14 T. Higashiguchi, H. Terauchi, N. Yugami, T. Yatagai, W. Sasaki, R. D’Arcy, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 96, 131505 (2010). 15 J. Sugar and C. Corliss, J. Phys. Chem. Ref. Data 14(Suppl. 2), 1 (1985). 16 J. E. Sansonetti, J. Phys. Chem. Ref. Data 37, 7 (2008). 17 T. Higashiguchi, H. Terauchi, T. Otsuka, M. Yamaguchi, K. Kikuchi, N. Yugami, T. Yatagai, W. Sasaki, R. D’Arcy, P. Dunne, and G. O’Sullivan, J. Appl. Phys. 109, 013301 (2011). 18 T. Higashiguchi, M. Yamaguchi, T. Otsuka, H. Terauchi, N. Yugami, T. Yatagai, R. D’Arcy, P. Dunne, and G. O’Sullivan, Appl. Phys. Lett. 98, 091503 (2011). 19 R. C. Spitzer, T. J. Orzechowski, D. W. Phillion, R. L. Kauffman, and C. Cerjan, J. Appl. Phys. 79, 2251 (1996). 20 D. Colombant and G. F. Tonon, J. Appl. Phys. 44, 3524 (1973). 21 T. Fujimoto, J. Phys. Soc. Jpn. 47, 265 (1979); 47, 273 (1979). 22 H. Griem, Principle of Plasma Spectroscopy (Cambridge Monographs on Plasma Physics, 2005). 23 T. Higashiguchi, K. Kawasaki, W. Sasaki, and S. Kubodera, Appl. Phys. Lett. 88, 161502 (2006). 24 T. Higashiguchi, N. Dojyo, M. Hamada, W. Sasaki, and S. Kubodera, Appl. Phys. Lett. 88, 201503 (2006). 25 N. Hirashita, M. Kinoshita, I. Aikawa, and T. Ajioka, Appl. Phys. Lett. 56, 451 (1990). 26 M. Wasamoto, M. Katto, M. Kaku, S. Kubodera, and A. Yokotani, Appl. Surf. Sci. 255, 9861 (2009).
This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitationnew.aip.org/termsconditions. Downloaded to IP: 141.212.109.170 On: Fri, 21 Nov 2014 18:14:04
