Indoor vs Outdoor Geophysics
Research in mineral physics is essential for interpreting observational data from many other disciplines in the Earth Sciences, from geodynamics to seismology to geochemistry to petrology to geomagnetism to planetary science, and extending also to materials science and climate studies. The field of high-pressure mineral physics is highly interdisciplinary. Mineral physicists do not always study minerals nor use only physics; they study the science of materials which comprise the Earth and other planets and employ the concepts and techniques from chemistry, physics, materials science, and even biology. Observations from geochemistry and geophysics studies lead to the development of petrologic, seismic and geodynamical models of the Earth's deep interior. The goal of mineral physics is to interpret such models in terms of variations of pressure, temperature, mineralogy/crystallography, and/or chemical composition with depth. The discovery in 2004 of the post-perovskite phase of MgSiO3 at pressures in excess of 120 GPa and high temperatures has led to an explosion of both complimentary experimental and theoretical work in mineral physics and remarkable synergy between mineral physics and the disciplines of seismology, geodynamics and geochemistry. Similarly, the observation of high-spin to low-spin transitions in Fe-bearing minerals at high pressures has important implications for the lower mantle of the Earth. We focus in this talk on the use of experimental physical acoustics to conduct "indoor seismology" experiments to measure sound wave velocities of minerals under the pressure and temperature conditions of the Earth's mantle. This field of research has a long history dating back at least to the studies of Francis Birch in the 1950s. The techniques include ultrasonic interferometry, resonant ultrasound spectroscopy, and Brillouin spectroscopy. Many of these physical acoustic experiments are now performed in conjunction with synchrotron X-radiation sources at national and international facilities. The role of mineral physics research is to provide experimental data and theoretical computations to allow interpretation of observational data from other branches of geosciences in terms of the chemical and thermal state of the Earth's interior and its evolution over geological time.
Geophysics Studies With High-Resolution X-ray Spectroscopy
The introduction of high-resolution (about 1 meV) inelastic x-ray scattering techniques to third generation synchrotron radiation facilities around the world has been very successful. New opportunities for the study of vibrational properties of condensed matter have emerged for research areas like biophysics, geophysics, and nanoscience. In particular, the determination of phonon dispersion relations with momentum-resolved inelastic x-ray scattering (IXS), of vibrational density of states with nuclear resonant inelastic x-ray scattering (NRIXS), and of the determination of valences, spin states, and magnetic ordering with synchrotron Mössbauer spectroscopy (SMS), all under extreme conditions, provided novel and often remarkable results in the scientific area of geophysics . In this contribution, the combination of high-resolution spectroscopy with diamond anvil cell technology and its impact on the geoscientific area will be discussed. Nuclear resonant spectroscopy (NRIXS & SMS) under extreme conditions has become a key method to provide sound velocities and elasticity on iron, iron alloys, and iron oxides, to study valence and high-spin to low-spin transitions in lower mantle minerals, and to investigate magnetic ordering transitions of iron-bearing materials. Its sensitivity in combination with isotope selectivity allowed investigations on materials under high pressures using diamond anvil cells and Laser heating. Examples will illustrate the present and potential future use of nuclear resonant spectroscopy in the Earth and planetary sciences. Momentum resolved IXS detects phonon dispersions and has been applied to Earth materials like iron metal and MgO under high pressure in diamond anvil cells. In single crystals, the dispersion of the acoustic phonons at low energies provides sound velocities for various directions and potentially the elastic tensor of the material. The technical requirements for nuclear resonant andmomentum resolved IXS methods have favored the diamond anvil cell for generation of high pressures. The x-ray energies are typically between 10 keV and 30 keV, and the high-pressure device has to accommodate sufficiently low absorption at those energies. In addition, momentum-resolved IXS needs access to a plane with 20-30 degree opening angle. NRIXS experiments require access to a significant solid angle. These conditions have been more readily met by diamond anvil cells. Here we will discuss the possibility of using a multi-anvil device instead and evaluate the conditions for IXS experiments in such apparatus. This work is supported by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. References  W. Sturhahn and J.M. Jackson, in Advances in High-Pressure Mineralogy, E.Ohtani, ed., Special Paper 421, 157-174 (2007)
Toward a self-consistent pressure scale: elastic moduli and equation of state of MgO and Ringwoodite by simultaneous x-ray density and Brillouin sound velocity measurements at high-pressure high-temperature conditions
Accurate phase diagrams and PVT equations of state (EOS) of materials strongly depend on the PVT calibrations of standard materials (e.g. MgO, NaCl, Au, Pt), which currently do not predict identical pressures at the same experimental conditions. MgO is commonly used as a pressure standard in a variety of high pressure and high-temperature experiments. Despite being one of the simplest and most studied materials, its accurate EOS is still uncertain, especially at high PT. The direct way of obtaining a self consistent pressure scale is by measuring acoustic velocities (Vp and Vs) and density simultaneously. Such P-V-T-Vp-Vs measurements allow one to determine the pressure directly, without resort to a separate calibration standard. Recently, as part of a major COMPRES initiative, we have constructed a Brillouin spectrometer at GSECARS, APS (13-BM-D) which allows accurate simultaneous sound velocity and lattice parameter measurements at high pressures and high temperatures. Such measurements were performed on single crystal MgO at simultaneously high pressures (up to 30 GPa) and high temperatures (up to 873K) in diamond cells. At each PT point we measured the unit cell parameters and the acoustic velocities of MgO in several crystallographic directions, and directly obtained all three single crystal elastic moduli, as well as isotropic adiabatic bulk (Ks) and shear (μ) moduli. Unit cell parameters of pressure medium (Ne, Ar) and additional pressure calibrants (Au, Pt, NaCl) were measured at each PT for cross calibration. In addition we demonstrate that successful P-V-T-Vp-Vs measurements can be performed on certain polycrystalline materials, e.g. Ringwoodite (γ-Mg2SiO4). The results of these experiments and implications for a self consistent P-V-T(-Vp-Vs) pressure scale will be presented and discussed.
Seismic velocities and anisotropy in subducting slabs: Constrains from high pressure Brillouin scattering studies on hydrous phases
Water transported and released into the upper mantle via subduction of oceanic lithosphere has a profound effect on the physical and mechanical properties of mantle materials and may trigger earthquakes and partial melting. The identification of water storage sites in the slab is therefore necessary to constrain H2O recycling through subduction zones and the effect that its circulation on a global scale has on the dynamics of the Earth's interior. As seismology represents the preferred method to detect hydration, knowledge of the sound velocities and elastic properties of candidate hydrous minerals are essential to interpret the seismic velocity structure and anisotropy of subducted plates. Dense hydrous magnesium silicates (DHMS) are recognized as important host for H2O in the slab, but their elastic properties under the appropriate pressure-temperature conditions are still poorly constrained. Here I present recent high-pressure Brillouin spectroscopy measurements to determine the sound velocities and single-crystal elastic properties of Fe-bearing phase A (phA) and phase E (phE), two DHMS that may transport water into the upper mantle and transition zone. Measurements were performed on samples compressed up to 16.5(2) GPa in the diamond-anvil cell. The results provide new insights into the behavior of hydrous minerals under subduction conditions and the possibility of identifying hydration through seismic observations. In both cases, the shear properties of the materials are important factors in the conclusions reached. The compressional (VP) and shear (VS) wave velocities of phA and phE are significantly lower than those of other phases in slab peridotite with whom they coexist. The new data is used with existing thermoelastic data to compute the density and seismic velocity structure of harzburgitic subducted slabs with various degrees of hydration at pressures corresponding to the upper mantle and transition zone. The results suggest that the seismic velocities of dry and water-saturated harzburgites (44.5 vol% phA) may be indistinguishable at upper mantle P-T conditions because of the increasing concentration of fast high- pressure orthopyroxene upon hydration. On the other hand, the effect of phE on the seismic velocities of the slab is more pronounced, suggesting that water may be resolved seismically if this phase accounts for more than 15 vol% of subducted harzburgites (2 to 3wt% bulk water). Combined observations obtained from the analysis of seismic parameters indicate that the presence of significant shear wave anisotropy (As), together with high VP/VS and Poisson's ratios, could be major diagnostic features for identifying these hydrous phases at depth (180-410 km) in cold subudcted slabs.
Experimental and Other Estimates of Outer Core Viscosity
More than 40 estimates of the viscosity of Earth's outer core were made prior to 1994 based on observation and theory. In the past 15 years, high pressure experiments providing direct measurement of the viscosity of core mimetic liquid alloys have emerged as new input to our knowledge of the rheological properties of the outer core. Most of the experimental results confirm the low viscosity values predicted by theory. When combined with new estimates of viscosity based on observations of inner core motions (super-rotation, nutation, torsional oscillation, translation), the resulting updated data set shows that the differences in order of magnitude of viscosity estimates continues to span an order of magnitude. The large range of new estimates of viscosity continues to signify the differences in quantity being determined. The higher viscosity values produced by observation of the real Earth are attributed to an effective or eddy viscosity whereas the lower viscosity values produced from experiment and theory are attributed to an intrinsic or molecular viscosity. A review of new viscosity estimates over the past 15 years will be given, including some new high pressure experimental data we have obtained on Fe-17wt%Si.
Viscosity of Earth's Outer Core
The viscosity of Earth's outer fluid core has been the subject of a wide range of estimates. Direct observations of its viscosity tend to be much larger than those found from the extrapolation of laboratory high pressure and temperature experiments. The extrapolations characteristically give values near that of liquid iron at atmospheric pressure, while direct observations are generally many orders of magnitude larger. An exception to this dichotomy results from the extrapolation of laboratory measurements by Brazhkin (1998) and Brazhkin and Lyapin (2000) using the Arrhenius activation model. Although this model is widely used on the assumption that the activation volume is independent of pressure, measurements show that it increases strongly with pressure, yielding an estimate of 102 Pa.s at the top of the fluid core and 1011 Pa.s at the bottom. Of course, such extrapolations are subject to large uncertainties. In this paper, we review direct observations of the viscosity, at the top of the outer core from the decay of the Free Core Nutations, which give 2,371± 1,530 Pa.s, and at the bottom from the reduction in splitting of the two equatorial translational modes of oscillation of the solid inner core, which give an average value of 1.247± 0.035× 1011 Pa.s. Encouraged by the closeness of the Arrhenius extrapolation of laboratory measurements to the direct observations, we use a differential form of the Arrhenius activation model to interpolate along the melting curve to find a viscosity profile across the entire outer core.
Compressional Wave Velocities and Structure of Fe-Ni-Si and Fe-Ni-S Melts to 1650°C
High-precision high-temperature ultrasonic interferometric technique was used to measure VP in three molten
Fe-Ni-Si alloys containing varying amounts of Fe (75-89 wt % Fe), 5 wt % Ni, and varying amounts of Si (6-20
wt %) in the temperature range 1190-1650°C and in the frequency range 9-13 MHz. The temperature
dependencies of VP for all the compositions are linear, and no dispersion is observed in the temperature and
frequency ranges of the experiment. However, increasing amounts of Si cause a decrease in VP and an
increase in the (dVP/dT) value. The results for the Fe-Ni-Si melts are also compared with those previously
obtained for Fe, Fe-Ni and Fe-Ni-S melts. Whereas in the Fe-Ni-S melts, addition of S was observed to cause
anomalous elastic behavior (+dVP/dT), the behavior in Fe-Ni-Si is normal (-dVP/dT). The differences in the
high-temperature elasticity are discussed in light of the structural studies of the two systems by fluorescence
X-ray absorption fine structure (XAFS)1325°C.
Experimental Reflection and Transmission Studies on Water Loaded Plates: Application to Measurement of the Biot Slow Wave
An unique laboratory ultrasonic system that employs a 'large' aperture ultrasonic transmitter (8 cm X 6 cm) and a near 'point' receiver is applied to the study of the reflectivity of plates of elastic, viscoelastic, and porous materials. The analysis of the results is improved by modelling of the transmitters wavefield. This is particularly important in the at high angles of incidence and near critical angles where nonintiuitive beam effects occur. This modelling adds greatly to the reliability of quantitative measures of velocity, ampltiude, and attenuation. The method was calibrated on a series of elastic (metals) and viscoelastic (PMMA acrylic) plates prior to application to the more complicated water saturated plate. A strong slow wave was observed and modelled. As well the first measurements of the variation of the reflected amplitude with angle of incidence from a liquid saturated porous surface was obtained. The results appear generally consistent with the theories developed by Biot and others regarding propagating waves through fluid saturated porous materials.
Ultrasonic Measurements and Multi-Anvil Devices
The Earth's mantle has a mass of about 4.08 * 1021 tons and represents 68 % of the total mass of the
Earth. It is only accessible by indirect methods, above all seismological studies. The interpretation of seismic
data from the Earth's deep interior requires measurements of the physical properties of Earth materials under
experimental simulated mantle conditions. The simulation of these in situ conditions require high pressure
techniques - diamond anvil cells (DAC), multi-anvil devices (MAD) and mostly synthetical samples. MADs are
more limited in pressure, but provide sample volumes 3 to 7 orders of magnitude bigger. They also offer small
and even adjustable temperature gradients over the whole sample. The bigger samples make anisotropy and
structural effects in complex systems accessible for measurements in principle. The measurement of both
elastic wave velocities have also no limits for opaque and encapsulated samples. The ultrasonic
interferometry allows the highly precise travel time measurement at a sample enclosed in a high-pressure
multi-anvil device. Under high pressure conditions the influence of sample deformation is so important that
ultrasonic interferometry requires the exact sample deformation measurement under in situ conditions using
synchrotron radiation. There is a promising way to increase the maximum pressure of multi-anvil devices by
multi-staging, i.e. implementation of additional sub-anvil set-ups resulting in a better distribution and limitation
of the stress inside the anvils. Contrary to the common opinion of overshooting the maximum crushing
strength most of the anvils fail in high pressure experiments due to the exceeding of the maximum tensile
stress as a result of the lateral deformation. We present recent techniques and results of elastic properties
measurements performed at different multi-anvil devices. That comprises standard-free pressure
measurements, transient experiments, multi-cycle and multi-staging experiments as well as melt studies.