High Energy Density Physics 5 (2009) 276282

Contents lists available at ScienceDirect

High Energy Density Physics

journal homepage: www.elsevier.com/locate/hedp

Bow shocks formed by plasma collisions in laser irradiated semi-cylindrical cavities

Jorge Filevich a, *, Michael Purvis a, Jonathan Grava a, Duncan P. Ryan b, James Dunn c, Stephen J. Moon c, Vyacheslav N. Shlyaptsev a, Jorge J. Rocca a, b
a b c

NSF ERC for Extreme Ultraviolet Science and Technology and Department of Electrical and Computer Engineering, Colorado State University, Fort Collins, CO 80523, USA Department of Physics, Colorado State University, Fort Collins, CO 80523, USA Lawrence Livermore National Laboratory, Livermore, CA 94551, USA

a r t i c l e i n f o
Article history: Received 26 February 2009 Received in revised form 3 April 2009 Accepted 3 April 2009 Available online 16 April 2009 PACS: 52.50.Jm 52.65.-y 52.70.-m 42.55.Vc Keywords: Soft X-ray lasers Plasma shocks Interferometry Plasma simulations

a b s t r a c t
The formation of shocks in plasmas created by short pulse laser irradiation (l 800 nm, I z 1 1012 W cm2) of semi-cylindrical cavities of different materials was studied combining visible and soft X-ray laser interferometry with simulations. The plasma rapidly converges near the axis to form a dense bright plasma focus. Later in time a long lasting bow shock is observed to develop outside the cavity, that is shown to arise from the collision of plasmas originating from within the cavity and the surrounding at walls of the target. The shock is sustained for tens of nanoseconds by the continuous arrival of plasma ablated from the target walls. The plasmas created from the heavier target materials evolve more slowly, resulting in increased shock lifetimes. 2009 Elsevier B.V. All rights reserved.

1. Introduction The collision and subsequent interaction of dense plasmas created by intense laser irradiation of cylindrical cavities are of interest for fundamental and practical reasons [13]. We have recently reported the study of semi-cylindrical cavity plasmas using soft X-ray laser interferometry and hydrodynamic simulations [4]. Aluminum and carbon plasmas were created by irradiating half-cylinder cavities at intensities of 1 1012 W cm2 with 120 ps duration optical laser pulses. The plasmas were interferometrically probed with 46.9 nm laser light to obtain electron density maps at different times throughout their evolution. The reduced refraction of the soft X-ray probe relative to an optical probe allows the mapping of the electron density in plasma regions with higher density gradients [5,6]. In this semi-cylindrical target geometry pressure gradients at the walls radially accelerate the plasma towards a location near the cavity axis, where it collides

forming a bright high density plasma focus with an electron density > 1 1020 cm3. During the study of these plasmas we also observed extreme ultraviolet plasma self-emission from a long and narrow arc outside the cavity (see Fig. 1), indicative of the presence of a bow shock. Bow shocks are of interests in astrophysics [7,8] and have been studied in the laboratory [9]. Herein we report the study of these shocks using interferometry and two-dimensional radiation hydrodynamic code simulations. Optical interferometry was used to complement soft X-ray laser interferometry in mapping the lower density regions where the shorter wavelength probe is insensitive. The combination of both interferometry techniques provides the ability to measure, for these particular plasmas, electron densities within the range from 5 1017 cm3 to 1 1020 cm3, with the highest value limited by probe beam refraction.

2. Experimental setup The plasmas were created using a Ti:Sapphire laser beam to heat 500 mm diameter semi-cylindrical grooves machined into

* Corrresponding author. E-mail address: jorge.levich@colostate.edu (J. Filevich). 1574-1818/$ see front matter 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.hedp.2009.04.003

J. Filevich et al. / High Energy Density Physics 5 (2009) 276282

277

The plasma was probed using one of the two probe wavelengths at a time. The experimental geometry is shown in Fig. 2(a). The cavity was irradiated at normal incidence with respect to the at front surface of the target while the probe beam propagated along the axis of the 1 mm long semi-cylindrical cavity. The plasma was imaged onto a CCD camera with a magnication of 20 using an f 20 cm lens. A narrow band lter centered at l 532 nm was used to reduce the plasma self-emission collected by the imaging system. The resolution of this imaging system determined by the 1090% rise on a knife edge image was w10 mm. Electron density maps were obtained from the interferograms by assuming that the plasma is uniform along the direction of propagation and that the index of refraction of the plasma is dominated by the free electrons [12]. Under these conditions the electron density can be directly obtained from the measured number of fringe shifts

Fig. 1. Time integrated extreme ultraviolet emission from a Cu plasma created by laser irradiation of a semi-cylindrical target.

1 mm thick slabs of different materials with intensities of w1 1012 W cm2. The 120 ps short duration of the heating laser pulse results in rapid deposition of laser energy, making it possible to study the evolution of the plasma without further laserplasma interaction. Studies were performed for carbon, aluminum, copper and silver targets. The Ti:Sapphire laser beam was focused into the groove forming a line focus of w470 mm FWHM and of w1.3 mm length. The position and intensity distribution of the line focus were monitored for every laser shot by imaging the reection of the beam off an optical at onto a CCD camera placed at a distance equivalent to that of the target. The plasmas were imaged onto an MCP/CCD (Multi Channel Plate/Charged-Coupled Device) detector system with 25 magnication using a spherical Sc/Si multilayer-coated mirror. The plasmas were probed using two different interferometers to map the high and low electron density regions. The rst is a MachZehnder interferometer that operates at l 46.9 nm [6] using diffraction gratings to split the beam of a table-top 46.9 nm Nelike Ar capillary discharge laser. In this laser a fast electrical discharge current pulse compresses the plasma generated in an argon-lled capillary tube to generate a population inversion and amplication in the J 01 3p3s line of Ne-like Ar [10]. The laser delivered pulses of w1 ns duration and w0.15 mJ of energy [11]. The temporal jitter between the capillary discharge laser probe and the plasma heating laser pulse was reduced to less than 2 ns by laser triggering the capillary discharge main spark gap. This short wavelength probe can access the densities present in the early stages of the evolution of the plasma, in particular the region of the high density plasma focus. The second interferometer, used to probe the lower density regions in the plasma, is a MachZehnder interferometer that operates at l 532 nm. The temporal resolution of this visible interferometer is determined by the relatively long pulse, w8 ns FWHM, of the frequency doubled Q-switched Nd:YAG laser probe used. This pulse duration is nevertheless sufciently short to probe the regions of the plasma with a relatively slow varying density. The probe laser and the Ti:Sapphire plasma heating laser were synchronized with a jitter of less than 1 ns. The target was positioned to intersect one of the arms of the interferometer by using motorized translation stages.

Fig. 2. (a) Schematic of the semi-cylindrical target showing the incident plasma heating laser beam (from right to left) and the direction of propagation of the soft X-ray laser probe beam (perpendicular to the page). (b) Soft X-ray laser (46.9 nm) interferogram of a copper plasma showing the dense plasma build up near the axis.

278

J. Filevich et al. / High Energy Density Physics 5 (2009) 276282

Fig. 3. Sequence of interferograms depicting the evolution of a Cu plasma. The probe beam wavelength used was 532 nm.

in the interferograms (Nf ne l=2 ncrit l where ncrit is the critical density of the plasma at wavelength l). For a l 1 mm long plasma, one fringe shift at 46.9 nm probe wavelength corresponds to w5 1019 cm3, while at 532 nm probe wavelength, one fringe shift corresponds to w4 1018 cm3. More detailed descriptions of the experimental setup used to create the plasma and of the soft X-ray laser probe can be found in recent publications [4,6,13]. 3. Experimental data Fig. 1 shows the time integrated extreme ultraviolet plasma emission distribution corresponding to a copper plasma. Two bright regions are visible. The brightest region is located close to the axis of the semi-cylinder where the expanding wall plasma converges and collides forming the plasma focus. The second is located close to the target wall at the bottom of the cavity. A third dimmer structure with the shape of an arc is seen to develop outside the cavity, far from the region directly heated by the laser.

It was noticed that this arc structure appears in the self-emission images when the width of the plasma heating laser line focus was wide, i.e. w470 mm FWHM, approaching the grooves width. In this case the wings of the heating beam illuminate the frontal at surface surrounding the 500 mm groove target creating a plasma. In contrast, the arc structure was not observed in plasmas generated by a narrower line focus (w350 mm FWHM). The dynamics of the denser regions of the plasma where large density gradients are present were mapped using the soft X-ray laser interferometer. Fig. 2(b) shows a soft X-ray laser interferogram of a Cu plasma obtained 7.6 ns after laser irradiation. The white line indicates the target surface position. The wall plasma converges into a small region near the axis where it collides to form a plasma focus identiable by a sharp increase in the electron density. The plasma focus develops as early as 1.7 ns, reaching densities higher than 1 1020 cm3 at 7.6 ns in agreement with simulations. The simulations predict that the electron temperature in the plasma focus region reaches 35 eV. The plasma in this region

J. Filevich et al. / High Energy Density Physics 5 (2009) 276282

279

Fig. 4. Electron density maps obtained from the interferograms in Fig. 3. The density scale is logarithmic.

is observed to remain dense well after the time of its build up. This increase in plasma density matches the region of strong localized emission of radiation seen in Fig. 1. A series of soft X-ray laser interferograms describing the early part of the evolution of similar plasmas created by irradiation of aluminum and carbon targets was reported in Ref. [4]. Fig. 3 shows a sequence of selected l 532 nm interferograms, that describe the evolution of the lower density regions of the Cu plasma. The interferograms clearly show the development of a quasi-stationary bow shock structure outside the cavity. This plasma structure starts as two narrow arcs close to the axis of the semi-cylinder. As the plasma evolves, the arcs are stretched in length and the distance and angle between them widens slowly. Simultaneously, the width of the shock is observed to increase with time. This continues until the shock fades after about 70 ns. Fig. 4 shows the electron density maps obtained from the interferograms under the assumption that the plasma is uniform along the axial direction. The density maps show that the shock reaches a peak electron density of w6 1018 cm3 at 20 ns. The rst three frames of Fig. 4, that illustrate the early stages of the evolution, show plasma expanding from the at frontal surface of the target.

Similar bow shock structures were observed following the irradiation of carbon, aluminum and silver targets, corresponding to plasmas with atomic numbers ranging from 6 to 47. As expected, the speed at which the shocks evolve was observed to be progressively slower as the atomic mass of the material increases. For all materials the shocks are observed to evolve from a curved to a straight shock front. The speed at which this transition occurs depends on the atomic mass. Fig. 5 shows interferograms and electron density maps of C plasmas, the lightest material probed. A signicantly faster plasma evolution is observed in which the shock fronts are already straight in the 5 ns frame. For the heavier materials, Al, Cu and Ag, the shock fronts are rst observed to be straight at 35, 50 and 75 ns, respectively. The absence of signicant radiation cooling contributes to the disproportionately faster evolution of the C plasma. 4. Simulations and discussion The plasmas were modeled in two dimensions using the threedimensional single uid radiation hydrodynamics code HYDRA [14]. HYDRA is an Arbitrarily LagrangianEulerian code capable of

280

J. Filevich et al. / High Energy Density Physics 5 (2009) 276282

Fig. 5. Sequence of interferograms and density maps for a C plasma. The density scale is logarithmic. The main observed difference respect to the Cu plasma in Fig. 3 is a faster evolution.

running in an assortment of hydrodynamic mesh strategies to avoid the mesh tangling that commonly occurs when modeling a converging plasma. Inverse Bremsstrahlung absorption was assumed to be the dominant laser deposition mechanism at our irradiation conditions. The equation of state was modeled using the Lawrence Livermore National Laboratorys LEOS library [15]. Radiation transport within the plasma was treated using multi-group diffusion techniques with 100 bins spanning photon energies between 1 and 3 KeV. Heat conduction was simulated using the conductivities of Lee and More [16]. The electron ux limiter was set to a value of 0.05 though, at our relatively small laser uxes, this parameter is not critical. Simulated electron density maps of the copper plasma are shown in Fig. 6. The density distribution in the focal region is in very good agreement with the density build up seen in the soft X-ray interferometer maps. However, the opacities used in HYDRA are not accurate for the plasma conditions in the shock. Best agreement with the experiment is observed when radiation is turned off at 1 ns. In this case, the simulations reproduce well the plasma evolution including the density within the shock region (w6 1018 cm3 for Cu at 20 ns after the laser irradiation).

The origin of the shock formation is well illustrated by the computed map of plasma velocity vectors shown in Fig. 7. The shock arises when the plasma that originated from the groove, after having converged on axis, expands and collides with the plasma that originated from ablation of the at target wall surrounding the groove. The collision re-directs the velocity of the side plasma to follow the contour of the central expanding plasma producing a localized increase in plasma density, temperature and degree of ionization. The continuous arrival of material creates a quasistationary shock wave [17]. Fig. 8 shows the computed temperature and mean degree of ionization distribution maps of the Cu plasma at 20 ns in its evolution. The maps show that in the shocked region the electron temperature increases by about 50 percent (w4 eV) and that the degree of ionization also increases. Computation of the ionion collision mean-free path following Braginskii [18] gives a value always less than 1 mm, signicantly smaller than the width of the shock, an indication that this is a collisional shock. Simulations for the Cu plasma show that when the at wall surrounding the semi-cylindrical groove is not directly irradiated by the laser or by the plasma self-emission the arc-shaped shock does not develop. The shock is strongest when both types of

J. Filevich et al. / High Energy Density Physics 5 (2009) 276282

281

Fig. 6. Simulated electron density maps of Cu plasma at times corresponding to the experimental data of Fig. 4. The density scale is logarithmic.

irradiation are present. Simulations show that in the case of the heavier targets plasma radiation plays a role in ablating the front wall material, which contributes to the shock formation. In contrast, simulations performed for C plasmas reveal that radiation does not contribute to the ablation of the front wall. In any case, it is the continued ablation of the wall by either radiation or heat conduction that sustains the shocks for several tens of nanoseconds. Comparison of the evolution of plasmas from the different materials shows that the main difference between them is the slower dynamics of the heavier element plasmas, resulting from their larger mass. Simulations for C and Cu plasmas show that close to the target wall the initial temperatures and pressures are similar, but the difference in mass causes the C plasma to evolve more rapidly. For C the shock is rst seen to form a few ns after laser irradiation and to reach its peak electron density at 10 ns. The computed temperatures at 5 ns and 10 ns are 14 eV and 5 eV and the degrees of ionization are Z 4 and Z 2.5, respectively. In contrast, the Cu shock is observed to develop later, and to reach its peak density at w35 ns where the electron temperature is 6 eV and the degree of ionization is Z 4. The peak temperature, 15 eV,

Fig. 7. Velocity vector eld map at 20 ns in the evolution of the copper plasma. The arrow colors assist in visualizing the magnitude of the velocity.

282

J. Filevich et al. / High Energy Density Physics 5 (2009) 276282

Acknowledgments The authors would like to thank M. Marinak for helpful discussions regarding the HYDRA simulations. This research was sponsored by the National Nuclear Security Administration under the Stewardship Science Academic Alliances program through U.S. Department of Energy Research Grant #DE-FG52-06NA26152, using facilities from the NSF ERC Center for Extreme Ultraviolet Science and Technology, award EEC-0310717. Part of this work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under contract DE-AC5207NA27344. The work of M. Purvis was partially supported by a fellowship from the Institute for Laser Science Applications.

References
[1] T.R. Dittrich, S.W. Haan, M.M. Marinak, S.M. Pollaine, D.E. Hinkel, D.H. Munro, C.P. Verdon, G.L. Strobel, R. McEachern, R.C. Cook, C.C. Roberts, D.C. Wilson, P.A. Bradley, L.R. Foreman, W.S. Varnum., Review of indirectdrive ignition design options for the national ignition facility, AIP 6 (1999) 21642170. [2] John Lindl, Development of the indirect-drive approach to inertial connement fusion and the target physics basis for ignition and gain, Physics of Plasmas 2 (11) (1995) 39334024. [3] J.E. Balmer, R. Weber, P.F. Cunningham, P. Ladrach., Plasma evolution in laserirradiated hollow microcylinders, Laser and Particle Beams 8 (12) (1990) 327337. [4] Mike Purvis, Jonathan Grava, Mario C. Jorge Filevich, James Dunn Marconi, J.Moon Stephen, Vyacheslav N. Shlyaptsev, Elizabeth Jankowska, Jorge J. Rocca., Dynamics of converging laser-created plasmas in semicylindrical cavities studied using soft X-ray laser interferometry, Physical Review E 76(4) (046402) (2007). [5] L.B. Da Silva, T.W. Barbee Jr., R. Cauble, P. Celliers, D. Ciarlo, S. Libby, R.A. London, D. Matthews, S. Mrowka, J.C. Moreno, D. Ress, J.E. Trebes, A.S. Wan, F. Weber., Electron density measurements of high density plasmas using soft X-ray laser interferometry, Physical Review Letters 74 (20) (1995) 3991. [6] J. Filevich, K. Kanizay, M.C. Marconi, J.L.A. Chilla, J.J. Rocca, Dense plasma diagnostics with an amplitude-division soft-X-ray laser interferometer based on diffraction gratings, Optics Letters 25 (5) (2000) 356358. [7] P. Hartigan, J. Raymond, L. Hartmann., Radiative bow shock models of Herbig Haro objects, Astrophysical Journal 316 (1987) 323348. [8] D. Vignes, C. Mazelle, H. Rme, M.H. Acuna, J.E.P. Connerney, R.P. Lin, D.L. Mitchell, P. Cloutier, D.H. Crider, N.F. Ness., The solar wind interaction with mars: locations and shapes of the bow shock and the magnetic pile-up boundary from the observations of the MAG/ER experiment onboard mars global survey, Geophysics Research Letters 27 (1) (2000) 4952. [9] A.R. Bell, P. Choi, A.E. Dangor, O. Willi, D.A. Bassett, C.J. Hooker., Collisionless shock in a laser-produced ablating plasma, Physical Reviews A 38 (3) (1988) 13631369. [10] J.J. Rocca, V. Shlyaptsev, F.G. Tomasel, O.D. Cortazar, D. Hartshorn, J.L.A. Chilla., Demonstration of a discharge pumped table-top soft-X-ray laser, Physical Review Letters 73 (16) (1994) 21922195. [11] B.R. Benware, C.D. Macchietto, C.H. Moreno, J.J. Rocca., Demonstration of a high average power tabletop soft X-ray laser, Physical Review Letters 81 (26) (1998) 58045807. [12] J. Filevich, J.J. Rocca, M.C. Marconi, S.J. Moon, J. Nilsen, J.H. Scoeld, J. Dunn, R.F. Smith, R. Keenan, J.R. Hunter, V.N. Shlyaptsev., Observation of a multiply ionized plasma with index of refraction greater than one, Physical Review Letters 94 (035005) (2005). [13] J. Filevich, J.J. Rocca, M.C. Marconi, R.F. Smith, J. Dunn, R. Keenan, J.R. Hunter, S.J. Moon, J. Nilsen, A. Ng, V.N. Shlyaptsev, T.W. Barbee., Picosecond resolution soft X-ray laser plasma interferometry, Applied Optics 43 (19) (2004). [14] M.M. Marinak, S.W. Haan, T.R. Dittrich, R.E. Tipton, G.B. Zimmerman., A comparison of three-dimensional multimode hydrodynamic instability growth on various national ignition facility capsule designs with [small-caps hydra] simulations, Physics of Plasmas 5 (4) (1998) 11251132. [15] D. Young, E. Corey. Release of new LEOS equation of state library. Lawerence Livermore National Laboratory unpublished memo, 1998. [16] Y.T. Lee, R.M. More., An electron conductivity model for dense plasmas, Physics of Fluids 27 (1984) 12731286. [17] V.A. Gribkov, Oleg N Krokhin, V. Ya Nikulin, O.G. Semenov, G.V. Sklizkov., Experimental investigation of nonspherical cumulative laser plasma congurations, Soviet Journal of Quantum Electronics 5 (5) (1975) 530537. [18] S.I. Braginskii., Transport processes in a plasma, , In: Reviews of Plasma Physics, vol. 1, Consultants Bureau, New York, NY, 1965, 205310.

Fig. 8. Temperature and degree of ionization maps corresponding to 20 ns in the evolution of the Cu plasma.

occurs at 8 ns when the degree of ionization is Z 8. The slower evolution of the plasma corresponding to the heavier target materials was observed to result in increased shock lifetimes: w40 ns for C, w50 ns for Al, w70 ns for Cu and w90 ns for Ag. 5. Conclusions The dynamics of plasma collisions and shock generation created by short pulse (120 ps) laser irradiation (I z 1 1012 W cm2) of semi-cylindrical cavities machined into at C, Al, Cu and Ag slab targets were studied combining visible and soft X-ray interferometry with hydrodynamic simulations. Measured and computed electron density maps agree well in describing the evolution of the plasma. The convergence of material ablated from the walls near the axis results in a plasma collision that forms a dense bright plasma focus, which reaches an electron density of w1 1020 cm3. Later in time, a quasi-stationary bow shock is observed to develop as a result of the collision between the expanding central plasma and plasma generated by ablation of the at walls that surround the semicylindrical groove. The shock, that is collisional in nature, reaches in the Cu plasmas a peak measured electron density of w6 1018 cm3 and at that time is characterized by an electron temperature of 6 eV and a degree of ionization of Z 4. The shocks are sustained for several tens of nanoseconds by the continuous arrival of plasma from the target walls. The slower dynamics of the plasmas corresponding to the heavier materials results in an increased persistence of the shocks, that in Ag are observed to last for w90 ns.