THE concept for x-ray lasers goes back to the 1970s, when physicists realized that laser beams amplified with ions would have much higher energies than beams amplified using gases. Nuclear explosions were even envisioned as a power supply for these high-energy lasers. That vision became a reality at the time of the Strategic Defense Initiative of the 1980s, when x-ray laser beams initiated by nuclear explosives were generated underground at the Nevada Test Site. Livermore's Novette, the precursor of the Nova laser, was used for the first laboratory demonstration of an x-ray laser in 1984. Since then, Nova, Livermore's largest laser, has set the standard for x-ray laser research and been the benchmark against which x-ray laser research has been measured. Nova uses a very-high-energy pulse of light about a nanosecond (a billionth of a second) long to cause lasing at x-ray frequencies. Because these high-energy pulses heat the system's glass amplifiers, Nova must be allowed to cool between shots. Nova can thus be fired only about six times a day. In contrast, a team at Livermore has developed a small "tabletop" x-ray laser that can be fired every three or four minutes. By using two pulses--one of about a nanosecond and another in the trillionth-of-a-second (picosecond) range--their laser uses far less energy and does not require the cooling-off period. Scientists had theorized for years that an x-ray laser beam could be created using an extremely short, picosecond pulse, which would require less energy. But very short pulses overheated the glass amplifiers, destroying them. Laser chirped-pulse amplification, developed in the late 1980s, gets around that problem by expanding a very short pulse before it travels through the amplifiers and then compressing it to its original duration before the laser beam is focused on a target.1 If chirped-pulse amplification is combined with lower energies, the pulses do not overheat the glass amplifiers, so the system can be fired many times a day. The development team for this new laser includes Jim Dunn, the experimentalist, and theoreticians Al Osterheld and Slava Shlyaptsev, a visiting scientist from Russia's Lebedev Institute. All are physicists in the Physics and Space Technology Directorate. Together, they have produced one of only a handful of tabletop x-ray lasers in the world (Figure 1). X-ray lasers produce "soft" x rays, which is to say their wavelengths are a bit longer than those used in medical x rays. Soft x rays cannot penetrate a piece of paper, but they are ideal for probing and imaging high-energy-density ionized gases, known as plasmas. X-ray lasers are an invaluable tool for studying the expansion of high-density plasmas, particularly laser-produced plasmas, making them useful for Livermore's fusion and physics programs. Basic research using x-ray lasers as a diagnostic tool can fine-tune the equations of state of a variety of materials, including those used in nuclear weapons and under investigation by the Stockpile Stewardship Program. These lasers also have applications for the materials science community, both inside and outside the Laboratory, by supplying detailed information about the atomic structure of new and existing materials. Notes Osterheld, "Plasmas do not behave nicely. To verify the modeling codes for plasmas, we need lots of experiments." With an experiment every three or four minutes on the tabletop x-ray laser, large quantities of data can be produced quickly. The team's goal is to refine the process and reduce the size and cost of the equipment so that someday an x-ray laser might be a routine piece of equipment in plasma physics research laboratories. Achieving a Stable Lasing Plasma In x-ray lasers, a pulse of light strikes a target, stripping its atoms of electrons to form ions and pumping energy into the ions ("exciting" or "amplifying" them). As each excited ion decays from the higher energy state, it emits a photon. Many millions of these photons at the same wavelength, amplified in step, create the x-ray laser beam. The highly ionized material in which excitation occurs is a plasma (which should not be confused with the plasma that the x-ray laser beam is later used to probe). X-ray lasers are specifically designed to produce a lasing plasma with as high a fraction of usable ions as possible to maximize the stability and hence the output energy of the laser. If the target is made of titanium, which has 22 electrons, the ionization process strips off 12 electrons, leaving 10, which makes the ions like a neon atom in electron configuration. Neonlike ions in a plasma are very stable, closed-shell ions. They maintain their stability even when faced with temporal, spatial, and other changes. Dunn, Osterheld, and Shlyaptsev have also studied palladium targets. When palladium atoms are stripped of 18 electrons, their ions become like a nickel atom, which is also closed-shell and stable. A One-Two Punch In Livermore's Nova laser, a high-energy, kilojoule pulse lasting a nanosecond or slightly less must accomplish three things: produce an initial line-focus plasma, ionize it, and excite the ions. Because the excitation, or heating, is happening relatively slowly compared to other plasma behavior, this process is called quasi-steady-state excitation. The tabletop x-ray laser is configured differently from Nova (Figure 2). It uses the compact multipulse terawatt (COMET) laser driver to produce two pulses. First, a low-energy, nanosecond pulse of only 5 joules strikes a polished palladium or titanium target to produce the plasma and ionize it. The pulse must accomplish less than the Nova pulse, so less energy is needed.

Then a 5-joule, picosecond pulse, created by chirped-pulse amplification, arrives at the target a split second later to excite the ions. Although the picosecond pulse uses 100 times less energy than a Nova pulse, its power is ten times higher because the pulse is one thousand times shorter. And its power density, which adds the length of the target to the power equation, is also very high. The brief, picosecond, "transient" plasma excitation plays a major role in the laser's effectiveness. During the ionization process, the plasma expands rapidly. In the quasi-steady-state approach used with Nova, excitation occurs while the plasma is continuing to expand and be heated so that much of the deposited energy is lost from the lasing process. With the transient scheme, excitation happens so fast that more ions in the plasma can contribute to the lasing. For plasma research purposes, the tabletop x-ray laser almost has it all--low energy requirements, high power, a repetition rate of a shot every four minutes, and a short wavelength. (Keep in mind that the shorter the wavelength of the laser, the more effectively it can penetrate high-density plasmas.) Two Plasmas in One Chamber To date, the Livermore team has studied neonlike titanium and nickel-like palladium transient schemes. It has produced the first transient-gain, nickel-like, x-ray lasing at 14.7 nanometers with a laser pump of less than 10 joules (Figure 3).2 The team is looking at various ways to maximize the laser's output, including using different target designs and delaying the arrival of the picosecond pulse to match the propagation of the x-ray laser in the gain region.

Within the next year, the team plans to have a second plasma in the target chamber. The first one will be for lasing, while the second will be studied and probed. The very-short-pulse x-ray laser probe will act as a strobe to "freeze" the action of the second plasma, resulting in clearer images of plasmas than any yet produced. And with an experiment every three or four minutes, there can be lots of excellent images. --Katie Walter

Key Words: chirped-pulse amplification, plasmas, soft x rays, tabletop x-ray laser.

References
1. For more information on chirped-pulse amplification, see Science & Technology Review, "Crossing the Petawatt Threshold," December 1996, pp. 4-11.
2. J. Dunn, A. L. Osterheld, R. Shepherd, W. E. White, V. N. Shlyaptsev, and R. E. Stewart, "Demonstration of X-Ray Amplification in Transient Gain Nickel-like Palladium Scheme," Physical Review Letters 80(13), 2825-2828 (1998).

For further information contact Jim Dunn (925) 423-1557 (dunn6@llnl.gov) or Al Osterheld (925) 423-7432 (osterheld1@llnl.gov).