Diamond anvil cells in 2006 reach 10- to 100-TPa (0.1–1 Gbar) pressure range with laser induced shock waves: Rich Murray 2011.01.17
http://www.pnas.org/content/104/22/9172.full free full text Achieving high-density states through shock-wave loading of precompressed samples Raymond Jeanloz * , † , ‡, Peter M. Celliers § , Gilbert W. Collins § , Jon H. Eggert § , Kanani K. M. Lee ¶ , R. Stewart McWilliams *, Stéphanie Brygoo ‖ , and Paul Loubeyre ‖ + Author Affiliations Departments of *Earth and Planetary Science and †Astronomy, University of California, Berkeley, CA 94720; §Lawrence Livermore National Laboratory, Livermore, CA 94550; ¶Department of Physics, New Mexico State University, Las Cruces, NM 88003; and ‖Commissariat à l'Energie Atomique, 91680 Bruyères-le-Châtel, France Edited by Ho-kwang Mao, Carnegie Institution of Washington, Washington, DC, and accepted March 7, 2007 (received for review September 19, 2006) Abstract Materials can be experimentally characterized to terapascal pressures by sending a laser-induced shock wave through a sample that is precompressed inside a diamond-anvil cell. This combination of static and dynamic compression methods has been experimentally demonstrated and ultimately provides access to the 10- to 100-TPa (0.1–1 Gbar) pressure range that is relevant to planetary science, testing first-principles theories of condensed matter, and experimentally studying a new regime of chemical bonding. high pressure planetary interiors diamond–anvil cell Hugoniot laser shock In nature, and specifically when considering planets, high pressures are clearly evident in two contexts: the conditions occurring deep inside large planetary bodies and the transient stresses caused by hypervelocity impact among planetary materials. In both cases, typical peak stresses are much larger than the crushing strength of minerals (up to ≈1–10 GPa, depending on material, strain rate, pressure, and temperature), so pressures can be evaluated by disregarding strength and treating the rock, metal, or ice as a fluid. Ignoring the effects of compression, the central (hydrostatic) pressure of a planet is therefore expected to scale roughly as the square of the planet's bulk density (ρ planet, assumed constant throughout the planet) and radius (R planet ): Here, the scaling factor is adjusted to match the central pressure of Jupiter-like planets (RJupiter and ρ Jupiter are the radius and bulk density of Jupiter, respectively), and the effects of compression and differentiation (segregation of dense materials toward the center of a planet) act to increase the central pressure for larger, denser, more compressed, or more differentiated planets relative to Eq. 1 . Consequently, peak pressures in the 1- to 10-TPa range exist inside large planets, with Earth's central pressure being 0.37 TPa and “supergiant” planets expected to have central pressures in the 10- to 100-TPa range. In addition to static considerations, impact (the key process associated with growth of planets and the initial heating that drives the geological evolution of planets) is also expected to generate TPa pressures. Impedance-matching considerations described below can be combined with Kepler's third law to deduce that peak impact pressures for planetary objects orbiting a star of mass M star at an orbital distance R orbit are of the order Scaling here is to the mass of the Sun, and the average density and orbit of Earth, the latter being in astronomical units (1 AU = 1.496 × 1011 m); also, the characteristic impact velocity (u 0) is taken as the average orbital velocity according to Kepler's law, u 0 = 2πR orbit /T orbit with T orbit being the orbital period, and Eq. 2 assumes a symmetric hypervelocity impact. While recognizing that materials have been characterized at such conditions through specialized experiments (e.g., shock-wave measurements to the 10- to 100-TPa range in the proximity of underground nuclear explosions and from impact of a foil driven by hohlraum-emitted x-rays) (1–3), laboratory experiments tend to achieve significantly lower pressures. As with planetary phenomena, both static (diamond-anvil cell) and dynamic (shock-wave) methods are available for studying macroscopic samples at high pressures, but these are normally limited to the 0.1- to 1-TPa range (4). Still, these pressures are of fundamental interest because the internal-energy change associated with compression to the 0.1-TPa (1 Mbar) level is roughly (5) with volume changes (ΔV) being ≈20% of the 5-cm3 typical molar volume of terrestrial-planet matter (here we consider a mole of atoms, or gram-formula weight, which is 3.5, 5, and 6 cm3 for diamond, MgO, and water, respectively, at ambient conditions). The work of compression thus corresponds to bonding energies (≈1 eV = 97 kJ per mole, characteristic of the outer, bonding electrons of atoms), meaning that the chemical bond is profoundly changed by pressures of 0.1 TPa. This expectation has been verified through numerous experiments showing that the chemical properties of matter are significantly altered under pressure: for instance, hydrogen, oxygen, and the “noble gas” xenon transform from insulating, transparent gas, fluids, or crystals at low pressure to become metals by ≈0.1 TPa (5, 6). In this article, we briefly describe laboratory techniques that have recently been developed for studying materials to the 10- to 100-TPa range of relevance to planetary science. In particular, as most planets now known are supergiants of several (≈1.5–8) Jupiter masses orbiting stars at distances of a fraction of 1 astronomical unit (7), Eqs. 1 and 2 imply a strong motivation for characterizing materials up to the 100-TPa (1 Gbar) level. To reach such conditions, we combine static and dynamic techniques for compressing samples: specifically, propagating a shock wave through a sample that has been precompressed in a diamond-anvil cell (Fig. 1). By starting with a material that is already at high (static) pressures, one reaches higher compressions than could be obtained by driving a shock directly into an uncompressed sample. Moreover, by varying the initial density (pressure) of the sample, and also by pulse-shaping the shock-wave entering the sample, one can tune the final pressure-density-temperature (P–ρ–T) state that is achieved upon dynamic loading. This tuning is particularly relevant to planetary applications, because the average temperature profile through the convective interior of a planet is isentropic, rather than following a shock-compression curve (Hugoniot). Precompression thus allows one to significantly reduce the heating that tends to dominate the highest-pressure dynamic experiments, which is important for better characterizing the interatomic forces under compression. Experimental Approach Diamond-cell samples are necessarily small, ≈100–500 μm in diameter by 5–50 μm in thickness, as it is the small area of the diamond tip (culet) that allows high pressures to be achieved. Shock compression of such small samples is not well suited to experiments involving mechanical impact, for example, by a projectile launched from a light-gas gun (which currently sets the state of the art for high-quality shock-wave measurements, but involves sample dimensions of centimeter diameter by millimeter thickness). Instead, a laser-generated shock wave is better suited to the dimensions of the diamond cell, with a well defined shock front of ≈200–500 μm diameter being readily achieved at presently available facilities. Several laser beams are typically focused onto the outer surface of one of the diamond anvils, so as to generate an intense pulse of light that is absorbed at the diamond surface (thin layers of laser-absorbing plastic and x-ray-absorbing Au usually are deposited on that diamond surface) (Fig. 2). The outermost diamond is thereby vaporized, launching a high-amplitude pressure wave into the anvil caused by a combination of the rapid thermal pressure generated in the diamond (resulting from heating at nearly constant volume) and linear-momentum balance (“rocket effect”) relative to the diamond vapor that expands outward, back toward the incoming laser beams. Such a high-amplitude wave has the property of being self-steepening for a material with a normal equation of state (∂Ks /∂P > 0 for the adiabatic bulk modulus Ks ). As a result, As a result, a shock front is created inside the anvil and propagates toward the sample (8, 9)....
