Ultra-High-Intensity Laser Physics at the LANL Trident Laser Facility

B.M. Hegelich, J.C. Fernández, J.A. Cobble, K.A. Flippo (P-24), B.J. Albright, E.S. Dodd, M.J. Schmitt (X-1), R. Perea (MST-7)
Excerpted from LA-14202-PR

Introduction

Modern ultra-high-intensity lasers are able to reach focal intensities of the order of 10¹⁸–10²¹ W/cm² where laser-plasma interactions become relativistic, and a variety of new effects emerge into a completely new regime of physics. These processes include relativistic self-focusing of the laser beam, which results in even higher intensities, laser-induced particle acceleration to MeV energies on a µm scale, x-ray lasers, laser-induced nuclear physics, and even the production of antimatter and other exotic particles. These effects can be applied to a variety of physics studies and can potentially be used for a number of applications. Concepts under consideration are next-generation accelerators, the jump-starting of inertial-confinement fusion with fast ignition, various medical applications, and laboratory astrophysics. One of the three beams of the Trident laser facility at LANL has been converted to deliver ultrashort laser pulses at the above-mentioned intensities so that researchers can participate in this exciting new field at the forefront of today’s physics. This beam is used to carry out a program that investigates the acceleration of ion bunches to MeV energies and MA currents and the interaction of these ions with different targets.

High-Irradiance Laser-Matter Interactions

An ultra-high-intensity laser pulse pointed at a solid target always interacts with a plasma because of its finite contrast ratio. Even in a laser with a relatively excellent contrast of ~ 10^-7 there will be a “pre-pulse” at 1–2 ns ahead of the main pulse with an intensity above 10¹² W/cm², which is high enough to instantly create a plasma. The main pulse will therefore always interact with a plasma and never with a solid target. Furthermore, at these intensities, the laser-plasma interaction is relativistic, i.e., the electrons gain energy on the order of their rest mass when moving in the electro-magnetic field of the laser pulse. The laser transfers energy to the electrons and accelerates them in the laser-propagation direction through the target to multi-MeV energies. This process is due to the v × B force from the magnetic component of the electromagnetic field, which becomes nonnegligible when v approaches the speed of light. As illustrated in Figure 1, the electrons will penetrate a thin foil target, exit out the back surface, and set up a virtual cathode—a very strong electric field, exceeding field strengths of a few 10¹² V/m (TV/m).

The electric field ionizes the rear surface and accelerates whatever ions are situated there to energies of many MeV. Protons have been accelerated to more than 60 MeV,¹ fluorine ions to above 100 MeV,² and lately high-Z palladium ions to 220 MeV.^3,4 In many beam parameters, those ion pulses now exceed those of conventionally accelerated ions by orders of magnitude, exhibiting pulse durations in the subpicosecond range, beam currents up to MA, and a transverse emittance ε_t < 0.001 π mm mrad.⁵ A typical conventional accelerator like the CERN Super Proton Synchrotron (SPS) has an emittance of ε_t < 1 π mm mrad. These parameters have rekindled interest in laser-accelerated ion beams for applications like proton radiography⁶, isochoric heating⁷, fast ignition⁸, and next-generation accelerators. The major difficulty for all these applications to date has been the large energy spread of the laser-accelerated ions, which typically exhibit a Maxwellian-like energy spectrum as shown in Figure 2. At LANL, we have demonstrated for the first time that quasi-monoenergetic ion beams can be generated by controlled target treatment before irradiating the metal foil target with an ultra-high-intensity laser. Furthermore, we also show the acceleration of a single charge state of one ion species directly in the forward target-normal direction. The accelerated C⁵⁺ ion bunch shown in Figure 3 exhibits a longitudinal emittance of ε_l < 2 × 10^-6 π eV s, exceeding that of conventional high-current accelerators by orders of magnitude. This new result shows the strong potential impact that ultra-high-intensity laser physics can have in many other areas of physics.

Accelerated Ions

Due to the vacuum conditions in ultra-high-intensity laser experiments, which typically are around 10^-6 mbar, all target surfaces are coated with water vapor and hydrocarbon layers, e.g., pump oil. That means that no matter what target material is used, the outer layer always contains protons. Because of its low ionization potential and because protons exceed every other ion’s charge-to-mass ratio by at least a factor of two, they are more efficiently produced and accelerated, drain the energy out of the acceleration process, and screen the accelerating electric field for the heavier particles. Accelerating other ions therefore requires the removal of the contaminating proton layers. This removal was demonstrated by Hegelich et al.² using the 100 TW laser at the École Polytechnique LULI research center with carbon and fluorine ions. To achieve our goal of mid- to high-Z ion acceleration at LANL’s Trident laser facility, we implemented the same kind of cleaning techniques in our ultra-high-intensity laser experiments. Trident delivers pulses of up to 30 J in as short as 600 fs, which corresponds to a power of ~ 30 TW. As such, Trident is currently the highest-energy subpicosecond laser in the U.S. The beam is aimed at an off-axis parabolic mirror within a vacuum chamber to focus the beam from its initial diameter of 6 in. down to a 20 µm spot on the target, achieving intensities in excess of 10¹⁹ W/cm². A sketch of a typical setup is shown in Figure 4.

To clean the target, we rely on two methods: Joule heating using either a strong direct current that is passed through the foil or a continuous-wave (cw) laser. Both methods are capable of heating the target to temperatures in excess of 1000 °C, which remove all hydrogen?-bearing contaminants. With no hydrogen present, remaining species on the rear surface are ionized by the electric field. The charge state with the highest charge-to-mass ratio is predominantly accelerated. Experiments on the LULI 100 TW laser and on the Trident laser successfully accelerated a wide range of low-Z ions to multi-MeV/nucleon energies. Figure 2(a) shows the spectra for helium-like beryllium, carbon, oxygen, and fluorine.

Moving from low-Z ions to mid- or high-Z ions proves to be more difficult. Although contaminants like water vapor can be cooked off by heating the target, heating cannot clean metal oxides, carbides, and nitrides on the surface—these contaminants usually have binding energies in the eV range. To overcome this problem we are working on two different approaches. The first approach is to use a second pulsed laser at relatively low intensity (~ 5 × 10¹⁰ W/cm²) to ablate the rear surface thus removing the contaminating oxides, etc. Although this approach worked in principle, the technical details are tricky and remain a subject of ongoing study. The second approach is to use a “magic” material that does not form oxides or other compounds. This approach is the easier solution, however, it limits the available target materials. One such “magic” material is palladium. With an atomic mass of 106, palladium (Z = 46) does not easily form oxides. As shown in Figure 2(b), we succeeded for the first time to accelerate a mid-Z material into a multi-MeV/nucleon-ion bunch using an ultra-high-intensity laser. In future experiments, we hope to improve this result, increasing the energy and the particle number and achieving greater control of the beam properties, e.g., ballistic focusing as shown in Figure 3. Once these goals are achieved, the palladium beam can be used to study isochoric heating in matter, effectively recreating in the laboratory conditions that are otherwise only found in the interior of large Jovian-like planets.

Monoenergetic Ions and Modeling

As mentioned earlier, we were the first to accelerate a monoenergetic MeV-ion bunch with a laser. When heating the palladium target to remove the hydrogen, a thin monolayer layer of carbon atoms remained at the rear surface. Because these carbon layers are very localized, all carbon atoms see the same field at the peak of the pulse and are ionized to the same charge state and instantly accelerated. As the field decreases after ~ 100 fs, all the carbon atoms are at the front of the ion expansion, in a position trailing the hot electrons, which effectively conserves their volume in phase space. The full width half maximum of 0.5 MeV per nucleon in the Thomson parabola spectrum shows a longitudinal emittance of this C⁵⁺ bunch smaller than 2 × 10^-6 eV seconds—about six orders of magnitude better than for the CERN SPS. We have used the one-dimensional hybrid code BILBO (Backside Ion Lagrangian Blow-Off) to help us understand this result. BILBO solves a Vlasov-Maxwell system analytically, calculating the boundary conditions for a nonlinear Poisson solver. The solver yields the electron density and electric fields and propagates the ions as kinetic particles. The code uses a threshold ionization model with atomic data for ionization energies of carbon and palladium and hot-electron-cooling models to account for the extraction of energy used to accelerate the ions. Using BILBO, we reproduced the measured spectra qualitatively (Figure 3) and seek to derive better models to understand and optimize beam production.

Conclusion

Future experiments will be directed towards better understanding of the underlying acceleration physics and towards the use of the accelerated ions for a number of different applications. Because of the modest integrated energy of the ions, they must be focused to one point, which can be achieved by using curved targets (Figure 5). The long-term goal of our project is to study transport and stopping mechanisms of high-current ion beams in cold, dense plasmas. These experiments require the use of Trident’s other beam lines to create the target plasma while the short-pulse beam generates the ion pulse. For these experiments and for other short-pulse physics applications, more energy in the short-pulse beam is desirable and in some cases even necessary. Therefore, the Trident short-pulse arm will be upgraded over the next 18 months to deliver pulses of ~ 115 J in less than 500 fs. This 200 TW upgrade puts Physics Division even more firmly at the forefront of modern science and opens new opportunities for programmatic research. It keeps Trident among the best of the existing short-pulse lasers in the world and makes even more exotic new physics accessible.

References

1. R.A. Snavely et al., “Intense high-energy proton beams from petawatt-laser irradiation of solids,” Physical Review Letters 85, 2945–2948 (2000).
B.M. Hegelich et al., “MeV ion jets from short-pulse-laser interaction with thin foils,” Physical Review Letters 89, 085002-1–085002-4 (2002);
[For nanosecond-laser work, resistive Joule heating was successfully accomplished by W. Ehler et al., “Effect of target purity on laser-produced plasma expansion,” Journal of Physics D: Applied Physics 13, L29–L32 (1980).]
B.M. Hegelich et al., “Spectral properties of laser accelerated mid-Z MeV/u ion beams,” in Proceedings of the 28th European Conference on Laser Interaction with Matter, Rome, September 6–10, 2004.
J.C. Fernández et al., “Laser-ablation treatment of short-pulse laser targets: Towards an experimental program on energetic-ion interactions with dense plasmas,” in Proceedings of the 28th European Conference on Laser Interaction with Matter, Rome, September 6–10, 2004.
T.E. Cowan et al., “Ultralow emittance, multi-MeV proton beams from a laser virtual-cathode plasma accelerator,” Physical Review Letters 92, 204801?]1–204801-4 (2004).
J.A. Cobble et al., “High resolution laser-driven proton radiography,” Journal of Applied Physics 92, 1775–1779 (2002).
P.K. Patel et al., “Isochoric heating of solid-density matter with an ultrafast proton beam,” Physical Review Letters 91, 123004-1–123004-1 (2003).
M. Roth et al., “Fast ignition by intense laser-accelerated proton beams,” Physical Review Letters 86, 436–439 (2001).

Acknowledgment

This work has been performed under the auspices of the U.S. DOE. We acknowledge support from 2004 Laboratory-Directed Research and Development/Directed Research Proposal 20040064 and experimental run time at Ecole Polytechnique, Paris, France.

For further information, contact Manuel Hegelich, 505-667-6989,
hegelich@lanl.gov.