1997 MRS Fall Meeting & Exhibit

Chairs

Russell Hemley, Carnegie Inst of Washington
William Nellis, Lawrence Livermore National Lab
Renata Wentzcovitch, Univ of Minnesota
Peter Yu, Univ of California-Berkeley

Symposium Support 

• Center for High Pressure Research,
• NSF Science and Technology Center
• General Electric Superabrasives
• Lawrence Berkeley National Laboratory of US-DOE
• Lawrence Livermore National Laboratory of US-DOE

Proceedings published as Volume 499 
of the Materials Research Society
Symposium Proceedings Series.

* Invited paper

SESSION DD1: EARTH MATERIALS AT HIGH PRESSURE 
Chair: Alexandra Navrotsky 
Monday Morning, December 1, 1997 
Staffordshire (W)

8:15 AM *DD1.0 
HIGH-PRESSURE TRANSFORMATION OF Al2O3. Nobumasa Funamori, Raymond Jeanloz, Geophysics, Univ. California, Berkeley, CA.

High-pressure x-ray diffraction indicates that ruby undergoes a transformation to the Rh2O3 (II) structure when heated at pressures above 100 GPa, in good agreement with predictions from ab initio quantum mechanical calculations. The high-pressure phase does not quench to ambient pressure, and the occurrence of this newly discovered phase transformation may affect the interpretations of both static (diamond-cell) and dynamic (shock-wave) experiments at ultra-high pressures.

8:30 AM *DD1.1 
ELASTICITY AND RHEOLOGY AT HIGH PRESSURE AND TEMPERATURE. Donald J. Weidner and Robert C. Liebermann, CHiPR and Department of Geosciences, SUNY, Stony Brook, NY.

Recent advances in synchrotron-based high pressure/high temperature studies enable several new sample characterizations in the 0 - 20 GPa pressure range and the 25 - 1500 C temperature range. Ultrasonic acoustic measurements of both compressional and shear velocities for single-crystal and polycrystalline samples have been carried out within this P/T range using interferrometric methods with frequencies between 10 and 40 MHz. The polycrystalline specimens are generally extremely well sintered with bench top acoustic velocities within experimental error of those predicted by single-crystal elasticity data. Synchrotron radiation is used to define sample volume and, with a standard, pressure while temperature is monitored with a thermocouple. Taken together, this data yields pressure derivative, temperature derivative, and cross-derivatives of the bulk and shear moduli. To date samples include periclase (MgO) and several high pressure minerals. Loose powders of polycrystalline samples exhibit several percent of strain broadening of the diffraction peaks upon pressurization. As samples are heated the time dependence of the peak narrowing serves as a metric for strength of the sample. Rheological properties of samples as strong as diamond and as weak as sodium chloride can be defined with such observations. In particular, the time dependence of the relaxation is characteristic of the effective exponent of a power law creep relation. This tool as applied to minerals can determine deviatoric stress as low as about 0.1 GPa. Data taken for several high pressure phases of common minerals suggest significant strengthening of the Earth with increasing depth due to the effects of the different phases.

9:00 AM *DD1.2 
PLASTIC DEFORMATION OF SOLIDS AT HIGH PRESSURES: SOME GEOLOGICAL APPLICATIONS. Shun-ichiro Karato, Univ. of Minnesota, Dept. of Geology and Geophysics, Minneapolis, MN.

Plastic deformation of solids under high confining pressure is relatively unexplored area of solid state physics that has important implications for the dynamics of planetary interiors as well as materials science itself. On the one hand, large confining pressure usually reduces the easiness of thermally activated motion of atoms and hence increases the resistance to plastic deformation in solids. On the other hand, at extremely high confining pressures, many materials including silicate minerals transform to metallic state in which plastic deformation would be relatively easy. High confining pressures often cause phase transformations which affect plastic properties through various mechanisms. Therefore, complicated effects of high confining pressures on plastic deformation is expected, but experimental studies have been rather sketchy due primarily to the difficulties in conducting quantitative deformation experiments under high pressure and temperatures. In this talk, I will review some of the experimental and theoretical studies on the effects of high confining pressure on high temperature deformation of solids with the emphasis on geological materials (namely silicates). Experimental techniques to investigate plastic deformation at high confining pressures include those involving moving unsupported pistons (the Griggs apparatus) and those involving the motion of supported pistons. Various merits and limitations of these techniques will be discussed with a brief description of a newly developed technique which allows us to investigate plastic properties at large strains up to P < 16 GPa and T < 2000 K. The application of this new technique to olivine yields an activation volume of V* 14 cm3/mol for dislocation creep (power law creep). This activation volume presumably corresponds to that for formation and/or migration of jogs (or kinks) on dislocations and is significantly smaller than that for diffusion (V* 6 cm3/mol). With this activation volume, the resistance to plastic deformation at 400 km depth in the Earth will be 104 times higher than that at atmospheric pressure compared at the same temperature.

9:30 AM *DD1.3 
BIFURCATION IN COMPLEX SOLIDS AT HIGH PRESSURES. Ross Angel, Bayerisches Geoinstitut, Bayreuth, GERMANY.

Advances in the precision and accuracy of high-pressure X-ray diffraction techniques have resulted in the identification of a growing number of complex solids which display bifurcation in their equilibrium P-T phase diagrams. That is, the phase diagram includes two structures with the same space group symmetry but which have distinct structural conformations and different thermodynamic properties. In general, the structures of these solids consist of anionic frameworks of relatively rigid polyhedra, with charge balance provided by relatively large cations of low-charge which occupy the cages formed by the frameworks. At ambient conditions the framework provides the cage sites with double- or multi-well potentials. At low temperatures and pressures the extra-framework cations each occupy one of the minima of the potentials; coupling between the sites (via the framework) leads to ordering, either ferro or anti-ferro in form. At high temperatures dynamic disordering of the cations leads to a phase transition to a higher-symmetry structure. Specific examples include the technologically important materials melilites, titanite and lead phosphate, as well as the mineral anorthite. In each of these materials in-situ diffraction studies have shown that the same symmetry change occurs at high pressures that is the same as that previously observed at high temperatures. In each case, however, the structure of the high-pressure phase appears to include a single-minimum potential for the cation within each cage. In addition, experimental studies of anorthite feldspar have shown that at simultaneous high temperatures and pressures there appears to be a rapid cross-over regime between the high-pressure structure with single-wells and the high-temperature structure with dynamic disorder of the extra-framework cations. Similar 'crossovers' or 'phase transitions' without symmetry change have also been observed in other complex materials at high pressures and ambient temperature, where they may represent the super-critical extension of a bifurcated phase diagram at lower temperature. In KTP there is a strong first-order phase transition at 6 GPa without symmetry change, while changes in compression mechanisms without symmetry change have been observed in the minerals orthopyroxene and microcline feldspar.

10:00 AM DD1.4 
ISOSTRUCTURAL VERSUS EQUILIBRIUM EQUATIONS OF STATE: IMPLICATIONS FOR MATERIALS AND EARTH SCIENCES. Robert M. Hazen, Joel Ita, and Hexiong Yang, Geophysical Laboratory and Center for High Pressure Research, Carnegie Institution of Washington, Washington, DC.

Pressure-volume equations-of-state (EOS) of crystalline solids impose important constraints on models of interatomic bonding, and they provide an essential foundation for interpreting seismic data from the earth's deep interior. Conventional tabulations of P-V-T EOS are compiled according to structure type and composition. Details on the state of order-disorder are usually omitted in discussions of EOS, even though ordering may be important in characterizing the thermochemistry and transport properties of minerals, alloys, ceramics, and other crystalline phases. Recent studies demonstrate that order-disorder phenomena may affect P-V-T EOS in two significant ways. First, many phases display a significant volume of disordering, (1). Silicate minerals commonly display up to 0.5%, and values exceeding 1% have been observed in oxides and sulfides. In addition, cation ordering may have a significant effect on elastic moduli. In the case of pseudobrookite-type , which displays a wide range of Mg-Ti cation order, bulk moduli for the fully ordered and disordered end members are 168 and 157 GPa, respectively - a 7% difference (2). This result suggests that cation order, in addition to composition and structure information, is important when documenting the elasticity of crystalline phases. EOS measurements for geophysically relevant materials are usually made assuming rapid and reversible P-V-T systematics. However, many experimental techniques measure P-V-T EOS over time scales too short to achieve an equilbrium ordered state. Because molar volume, compressibility, and presumably thermal expansivity are functions of the state of order, the resulting laboratory ``isostructural EOS will, in general, differ from the ``equilibrium EOS observed in nature. Determination of EOS of minerals relevant to mantle conditions must thus be performed in situ, on crystals that have ordered states equilibrated with respect to both temperature and pressure.

10:45 AM *DD1.5 
AB INITIO STUDIES OF HIGH PRESSURE SILICATE AND IRON PHASES AND THEIR GEOPHYSICAL SIGNIFICANCE. G. David Price, Department of Geological Sciences, University College London, London, UNITED KINGDOM.

The physical properties of silicate minerals and of iron and its alloys are of considerable interest to both material scientists and geophysicists alike. Geologically, the properties of silicate minerals determine the dynamics of the Earth's interior, which in turn determines the distribution and occurrence of earthquakes, volcanoes, etc, where as solid and liquid iron are fundamental to our understanding of the behaviour of the Earth's core, including the generation of the dynamo causing the Earth's magnetic field. The physical properties of minerals at the extreme temperatures (>4000 K) and pressures (up to 300 GPa) to be found in the Earth's interior are poorly understood, because of the obvious experimental limitations imposed by these conditions. Computer simulations, however, can provide an accurate means of calculating the thermoelastic and defect properties of materials at high pressures (P) and temperatures (T) via a variety of techniques. We will review recent developments, involving both ab initio and more approximate methods, in the description of the elastic and defect properties of major Earth-forming phases such as the polymorphs of Mg2SiO4 and MgSiO3-perovskite, which forms 40% by volume of our entire planet. We will also describe recent pseudopotential calculations on the high pressure phases of crystalline Fe, and will present the results from our quantum mechanical molecular dynamics simulation of liquid Fe at two high P,T states representative of the boundaries of the Earth's outer core with the inner core and the lower mantle.

11:15 AM *DD1.6 
FIRST PRINCIPLES INVESTIGATION OF THE ELASTICITY OF EARTH MATERIALS AT HIGH PRESSURE. Lars Stixrude, Dept. of Geological Sciences, University of Michigan, Ann Arbor, MI.

The elastic constants of earth materials provide a link between seismological observations of the otherwise inaccessible interior and its composition, temperature, mineralogy, and dynamics. Until recently, we have had no knowledge of the elastic constants of important minerals at the relevant pressures (3-140 GPa, corresponding to depths of 100-2890 km). Recent advances in experiment and theory have begun to fill this gap, providing new windows into the workings of the earth's interior. The recent development of first principles variable cell-shape molecular dynamics techniques make it possible to investigate from first principles the high pressure behavior and elasticity of the relatively complex structures that are important for understanding the earth's mantle. We will discuss the behavior of two minerals of the same composition, Mg2SiO4 but different structures: olivine and spinel. The results are discussed in terms of the full elastic constant tensors of these phases, structural compression mechanisms, seismic (acoustic) velocities and their elastic anisotropy. The elasticity of the two minerals, based on pseudo-hcp and pseudo-fcc packing of oxygens, respectively, show very different behavior at high pressure. While the elastic anisotropy of olivine depends weakly on pressure between 0 and 25 GPa, the anisotropy of spinel decreases with pressure initially, vanishing at 20 GPa before increasing again at higher pressure. This unusual behavior is understood in terms of a change of sign of the combination of elastic constants c11-c12-2c44, and a resulting interchange of fast and slow directions of acoustic wave propagation. The implications of these results for our understanding of the composition, mineralogy and viscous flow in the earth's mantle will be discussed.

11:45 AM *DD1.7 
MAGNETIC COLLAPSE AND THE BEHAVIOR OF TRANSITION METALS OXIDES AT HIGH PRESSURES. Ronald E. Cohen, Carnegie Institution of Washington and Center for High Pressure Research, Washington, DC; Igor Mazin, George Mason University, Fairfax, VA; D. G. Isaak, Institute of Geophysics and Planetary Physics, University of California, Los Angeles, Los Angeles, CA, and Azusa Pacific University, Azusa, CA.

Magnetic transitions caused by loss of local magnetic moments are generally expected in transition metals and transition metal compounds with increasing pressure due to increasing band widths with pressure. In some cases transitions are close to continuous, and in others they are predicted to be strongly first-order with possibility of hysteresis and metastability. We have studied magnetic collapse in in the transition metal monoxides FeO, NiO, CoO, and MnO as well as in more complex materials and in transition metals such as Fe using the LDA and GGA with the LAPW and LMTO methods (Science, 275, 654-657, 1997). Phase stability, including the relative stability of B1, B2, and B8 structures will be discussed.

SESSION DD2: DYNAMIC COMPRESSION 
Chair: William J. Nellis 
Monday Afternoon, December 1, 1997 
Staffordshire (W)

1:30 PM *DD2.1 
SHOCK TEMPERATURES, SOUND VELOCITIES, AND MELTING OF IRON. Thomas J. Ahrens, Kathleen G. Holland and George Q. Chen, California Institute of Technology, Pasadena, CA.

Previous iron shock temperature measurements at Caltech and LLNL relied on theoretical models of R, ratio of thermal diffusivities of iron and transparent anvil materials, LiF, sapphire, and diamond. New data where R is determined using temperature versus time pyrometry on 200 to 1000 Angstrom films of iron yield higher ratio (3 and 10) for R for sapphire at 165 and 240 GPa than calculated from electron gas and Debye theories. For LiF, R is 6 (experimentally) versus 18 (theoretically) at 166 GPa. Revised shock temperatures yield 5900 350 K for melting of epsilon (or a closely strcturally related phase) at 243 GPa in good agreement with McQueen and Brown's 1986 prediction. New data of sound speeds in iron, preheated to 1573K and shocked to 80 GPa indicate melting of gamma iron at 3000 300 K and 71 2 GPa. This is in approximate agreement with our (Chen and Ahrens, 1997) thermodynamic calculations, and Boehler's (1993) and Saxena et al.'s (1994) data.

2:00 PM *DD2.2 
LASER SHOCK PROCESSING OF METALS TECHNIQUES AND LASER TECHNOLOGY. Lloyd A. Hackel, C. Brent Dane, Lawrence Livermore National Laboratory, Livermore, C A.

We have developed a high energy (25J to 100J), high average power (500W) Nd:glass laser technology for use in generating intense and deep residual stresses in metals by means of an inertially confined shock. We generate shock pressures in the range of 10 kBars to 30kBars by condensing the laser fluence to approximately 200 J per square centimeter onto a carbon based coating at the metal surface. A tamping material covers the metal surface, transmitting the laser light while confining the plasma and allowing the shock pressure to appropriately build. Residual compressive stresses with surface intensities of up to 160 ksi are developed in alloys such as Inconel and Ti-6A1-4V and made to extend compressive up to depths exceeding 1 mm. Our unique laser technology has an average power capability of 20 to 50 times any commercially available system and allows for the first time consideration of a high throughput industrial laser shock peening process.

2:30 PM *DD2.3 
SHOCK-INDUCED DEFECTS IN BULK MATERIALS. George T. Gray III, Los Alamos National Laboratory, Los Alamos, NM.

It is thirty years since Dieter(1961) and over twenty years since Zukas(1966), Doran and Linde(1966), Mahajan(1970), and Leslie(1973), respectively presented reviews of the metallurgical effects of shock-wave effects on metals and alloys. In their reviews the systematic changes produced by the passage of shock waves on the subsequent structure / property behavior of materials were broadly presented. In each paper the microstructure / mechanical property changes correlated with known shock compression characteristics (peak pressure, duration, etc.) and the shock-induced defects produced in numerous metals and alloys were compared with their deformation behavior at ordinary rates of deformation. Research on a wide range of shock-induced effects on materials response has suggested that a fruitful framework for the construction of such a formalism is that based on well-established dislocation kinetics theory. This approach, based on the classic work of Kocks, Argon, and Ashby(1975), has been shown to provide a physically-solid basis upon which to model metal plasticity under more traditional loading paths. In this review the basic ideas of how seemingly disparate shockinduced effects in different material classes might be unified using such an approach are discussed. Details of the manner in which differences in the quite different shock response between fcc and bcc or hcp metals and alloys can be rationalized will be presented. Conceptual ideas of how loading rate, pulse duration, and release phenomena affect material shock response related to dislocation generation/storage, deformation twinning, and shock-induced phase transition concepts will be discussed.

3:30 PM *DD2.4 
DYNAMIC PRESSURES IN POROUS MATERIALS: MECHANICAL MODELING AND MATERIAL APPLICATIONS. Vitali F. Nesterenko, University of California, La Jolla, CA.

The different micromechanical models for dynamic loading of porous materials are analysed; their predictions, restrictions, and usefulness in determining the final properties of compacts after dynamic loading are compared. Some examples of novel materials applications involving dynamic loading are presented including: amorphous alloys, rapidly solidified powders, diamond synthesis from fullerens, high-Tc ceramics, submicron, nanocrystalline Zr2O3(Y2O3) ceramic powders and silicide formation as a result of shear induced chemical reactions.

4:00 PM *DD2.5 
HIGH PRESSURE CARBON BEHAVIOR INDUCED FROM CARBIDE HUGONIOTS. Toshimori Sekine, Eiichi Takazawa, Takamichi Kobayashi, National Inst for Res in Inorganic Materials, Tsukuba, JAPAN.

It is very interesting to investigate carbon at ultra-high pressures. Although several simulations on the high pressure carbon indicate the presence of post-diamond phase, currently the attainable pressure is still not enough to look at the high-pressure form of carbon experimentally. We present some results on carbon behavior induced from carbide Hugoniots which have been determined recently by us. SiC transforms to a NaCl type above 100 GPa and TiC displays a NaCl type structure up to 200 GPa. The molar volume of carbon in carbides can be accomodated in tetrahedral or octahedral site is obtained and compared with that of diamond as a function of pressure.

4:30 PM *DD2.6 
EQUATION OF STATE OF WARM CONDENSED MATTER. Troy W. Barbee III, David A. Young, Lawrence Livermore National Lab, Dept of Physics and Space Technology, Livermore, CA.

Recent advances in computational condensed matter theory have yielded accurate calculations of properties of materials. These calculations have, for the most part, focused on the low temperature (T=0) limit. An accurate determination of the equation of state (EOS) at finite temperature also requires knowledge of the behavior of the electron and ion thermal pressure as a function of T. Current approaches often interpolate between calculated T=0 results and approximations valid in the high T limit. Plasma physics-based approaches are accurate in the high temperature limit, but lose accuracy below TTFermi. We seek to ``connect up'' these two regimes by using finite temperature methods (including linear-response1 based phonon calculations) to derive an equation of state of condensed matter for TTFermi. We will present theoretical results for the principal Hugoniot of shocked materials, including carbon and aluminum, up to pressures P>100 GPa and temperatures T>104K, and compare our results with available experimental data.

SESSION DD3: POSTER SESSION: 
HIGH PRESSURE 
Chair: John V. Badding 
Monday Evening, December 1, 1997 
8:00 P.M. 
America Ballroom (W)

DD3.1 
SINTERING OF COMPACTS FROM NANOCRYSTALLINE DIAMONDS WITHOUT SINTERING AGENT. Eugene Ekimov, Adam Witek,* Bogdan Palosz,* Vladimir Filonenko, Alexander Gavriliuk, Valery Gryaznov, Stanislaw Gierlotka,* and Svetlana Stelmakh,* Institute for High Pressure Physics RAS, Troick, RUSSIA; *High Pressure Research Center, UNIPRESS, Warsaw, POLAND.

Additive free compacts of polycrystalline diamond were made in toroid high pressure camera under pressure 8 GPa and temperature from 800 to 2150oC. Commercial nanocrystalline diamond powder DALAN was used with grain sizes in the range from 8 to 20 nm. The diamond powder was placed in a cylindrical graphite heater and covered from top and bottom by Ti or BN gaskets. Annealing from 30 sec. up to 0.5 h was applied, 6 min. period was used as standard. The compacts were examined by X-ray diffraction and SEM methods; density (helium pycnometry), electrical and thermal conductivites and Vickers hardness were measured. The structure of starting and sintered compacts exhibits strong one dimensional disordering, typical for polytype materials like SiC. It was found that, microstructure of sintered compacts is strongly correlated with concentration of oxygen (a part of starting powders was annealed in 400oC in vacuum 10-5 torr). In general, there is no increase of size of the nanocrystalline grains in compacts obtained up to 2000oC. There is a general tendency of decrease of density of compacts with increase of sintering temperature what obviously results from graphitization; threshold of graphitization lies in the range 1000 to 1400oC. Typically, the nanocrystalline compacts sintered for 6 min. have density around 80 of theoretical value. The Vickers microhardness is not less than 22 GPa. The measured electrical conductivity is typical for insulators, the thermoconductivity is near the value of type IIA polycrystalline diamond. Preliminary results of sintering of nanocrystalline diamond with addition of silicon as sintering agent will be presented.

DD3.2 
HIGH PRESSURE POLYMORPHISM IN SILICA. David M. Teter and Russell J. Hemley, Geophysical Laboratory and Center for High Pressure Research, Carnegie Institution of Washington, Washington, DC.

The nature of silica under pressure has been of long and continuing interest due to its wide ranging implications in geophysics, materials science, and fundamental physics. We have derived over 100,000 silica structures using a novel method based upon fundamental crystal chemistry and used first-principles total-energy calculations to examine the most favorable structures. We find that an essentially infinite class of energetically competitive phases can be generated from arrays of oxygen with silicon occupying one-half of the octahedral sites. Calculations for specific structures provide an explanation for a number of recent high-pressure results for crystalline silica and allow us to understand the nature of the short- and intermediate-range order in the high-pressure amorphous state. Finally, we apply variational cluster expansion methods to study the possibility for order-disorder transitions in silica at high temperatures and pressures.

DD3.3 
THE PHASE DIAGRAM OF CARBON AT HIGH PRESSURE AND TEMPERATURE. James N. Glosli, Francis H. Ree, Lawrence Livermore National Laboratory, Livermore, CA.

The phase diagram of carbon at high temperature and pressure is explored via molecular dynamics using the bond order potential developed by Brenner1. The potential predicts correct structures for all the common hybridization states (sp, sp2 and sp3) of carbon and allows for the breaking and formation of covalent carbon bonds. The particular features examined are the graphite-diamond line, the diamond melting line, the graphite melting line and the local structure of the liquid.

DD3.4 
MECHANISM OF THREE-DIMENSIONAL POLYMERISATION OF FULLERITE C60 UNDER PRESSURE. Alexander Lyapin, Vadim Brazhkin, Svetlana Popova, Institute for High Pressure Physics RAS, Troitsk, Moscow, RUSSIA; Sergey Lyapin, Clarendon Laboratory, Physics Department, University of Oxford, Oxford, UNITED KINGDOM.

We have synthesised a set of crystalline phases, corresponding to different stages of the 3D polymerisation, by heating C60 to different temperatures under various pressures from 8 to 12.5 GPa. The structure of polymer phases was identified as fcc, the lattice parameter a decreasing with the synthesis temperature. The analysis of the intensity of x-ray crystalline peaks showed that the pristine C60 units retain their spherical symmetry for fcc phases with lattice parameters in the range 12.314.17 (the first stage of polymerisation), while the distortion of C60 molecules occurs for lower lattice parameter values (the second stage). We propose a model for the first stage of the 3D polymerisation. The model considers lattice parameter of fcc phases as continuous function of the share of covalently bonded molecular pairs. Bonding in 3D-polymerized fullerite crystal may be described as a solid solution of van der Waals and covalent intermolecular bonds. Raman spectra of strongly polymerised fcc phases (13 ) display two broad bands (200-800 and 1200-1700 cm-1) and look like amorphous spectra. For slightly polymerised phases (a=13.5 ) we observed both broad amorphous-like bands and narrow lines corresponding to the intramolecular modes of C60. In the framework of the model proposed mechanical characteristics of 3D-polymeryzed fullerite phases may be considered in terms of rigidity percolation. We have found that lattice parameter value 13.7 is the threshold for formation of rigid 3D-polimers of fullerite. Rigidity percolation can be considered as criterion for 3D polymerisation in general case. From analysis of the dependence for various pressures we have established that a crossover between 2D and 3D polymerisation of C60 occurs in the range 7-8 GPa.

DD3.5 
HIGH PRESSURE STUDY OF PHASE TRANSITIONS, AMORPHIZATION AND HYDROGEN BONDING IN DIVALENT HYDROXIDE COMPOUNDS: ZN(OH)2, CA(OH)2, SR(OH)2. Shirley Ekbundit, Kurt Leinenweber, and George H. Wolf, Arizona State University, Dept of Chemistry and Biochemistry, Tempe, AZ.

Spectroscopic studies of hydroxide compounds at high pressure have only been explored very recently. After Kruger reported a pressure-induced amorphization in Ca(OH)2, several studies on other isostructural hydroxides were carried out in order to further understand the high pressure behavior of these compounds. In these studies, Ca(OH)2, Co(OH)2, and Ni(OH)2 are found to undergo a pressure induced crystal-to-glass transformation at nearly the same pressure of 12 GPa. Furthermore, hydrogen bonding is induced as a result of increasing pressure. A recent study by Ekbundit et.al. has shown that for Ca(OH)2, the process of pressure-induced amorphization depends greatly on the grain size of the sample and only occurs for fine grained samples. For larger crystals of Ca(OH)2, a crystal-to-crystal phase transformation was found prior to the amorphization, which was postponed to 20 GPa. The high pressure structure of Ca(OH)2 was proposed to be that of Sr(OH)2, which increase the coordination number of calcium from six to seven. This follows a general trend of increasing coordination number with pressure. To investigate possible structural trends that can exist in the divalent hydroxides, we expand the system to Zn(OH)2, and Sr(OH)2, which crystallize in the orthorhombic system. The coordination number is four for zinc, six for calcium and seven for strontium. In the case of Sr(OH)2, amorphization was observed between 10-12 GPa followed by a phase transformation upon heating the sample, similar to the behavior seen in Ca(OH)2. The Zn(OH)2 system is currently under investigation. The amorphization, structural trends and the nature of hydrogen bonding in these compounds will be discussed.

DD3.6 
MOLECULAR DYNAMICS OF SHOCK LOADING OF METALS WITH DEFECTS. James Belak, University of California, Lawrence Livermore National Laboratory, Livermore, CA.

The finite rise time of shock waves in metals is commonly attributed to dissipative or viscous behavior of the metal. This viscous or plastic behaviour is due to the motion of defects such as dislocations. Despite this intuitive understanding, the experimental observation of defect motion or nucleation during shock loading has not been possible due to the short time scales involved. Molecular dynamics modeling with realistic interatomic potentials can provide some insight into defect motion during shock loading. However, until quite recently, the length scale required to accurately represent a metal with defects has been beyond the scope of even the most powerful supercomputers. Here, we present simulations of the shock response of single defects and indicate how simulation might provide some insight into the shock loading of metals.

DD3.7 
ELECTRICAL CONDUCTIVITY OF LOW-DENSITY SILICA AEROGEL AT 100 GPa (1 Mbar) PRESSURES. M. Bennahmias and W.J. Nellis, LLNL, Livermore, CA.

Electrical conductivity measurements were performed on low-density silica aerogel at pressures in the 100 GPa range using a two-stage light-gas gun. The initial density of the silica aerogel samples was 0.12 g/cc. The target geometry in these experiments was similar to that used previously for liquid hydrogen.[1] The techniques will be described and results of these measurements will be reported as a function of pressure and temperature.

DD3.8 
ELECTRICAL CONDUCTIVITY OF SHOCK-COMPRESSED SULPHUR, IODINE, BROMINE AND WATER. Vladimir V. Yakushev, Institute of Chemical Physics in Chernogolovka RAS, Moscow Region, RUSSIA.

Experimental investigation of sulphur, iodine, bromine and water electrical conductivity under shock compression are reviewed. Most reported experiments have been performed in the author's laboratory of the Institute of Chemical Physics RAS. We used different regimes of loading: single shock wave, double and multi step shock compressions. Pressure more than 50 GPa was realized by two plan convergent shock waves. Sulphur was found to transform into a metal state at approximately 30 GPa. Character of the transition was identified from the qualitatively temperature dependencies of the conductivities below and above this pressure, obtained from the comparison of single and double shock data. The value of sulphur resistivity was approximately 0.001 ohm cm at 30 GPa and drop slowly with further pressure rise. Similar results have been obtained for iodine. Its resistivity run to 0.001 ohm cm at approximately 15 GPa and increased moderately up to 110 GPa. An intriguing problem is the nature of liquid bromine conductivity. It was shocked in a teflon sell with platinum electrodes. Resistivity at 9 GPa was found to be 0.08 ohm cm and approximately 0.01 ohm cm at 30 GPa. Unexpected results were obtained in electrochemical experiments in which galvanic cells with copper and aluminium electrodes and bromine as electrolyte were subjected to shock loading and their electrical response was measured. We discovered that the sells produce open circuit voltages of about 0.5 V under shock pressures from 7.4 to 30 GPa. This voltage level clearly shows that the observed electrical signals are electrochemical in origin and connected with an ionic character of bromine conductivity. Analogous experiments were made with water. Ionic resistivity at 75 GPa was found to be 0.05 ohm cm.

DD3.9 
SUPERCRITICAL FLUID PHASE SEPARATIONS INDUCED BY CHEMICAL REACTIONS*. Francis H. Ree, Lawrence Livermore National Laboratory, Livermore, CA.

Currently available studies on supercritical fluid phase separations are limited to chemically nonreactive systems. The present study is concerned with a possible influence of chemical reactions on such a phase change. Our statistical mechanical study predicts that chemically reactive systems containing species composed of C, H, N, O, F atoms can exhibit a phase separation in which F atoms mainly appear as a constituent of HF in a N2-rich fluid phase up to a certain pressure beyond which they occur as CF4 in a N2-poor phase. The pressure at the phase boundary can be about 10 GPa to 20 GPa at 1000 K. It may be accessible by present-day experimental high-pressure techniques. We discuss implications of this prediction on the performance of high explosives.

DD3.10 
STRUCTURAL PRESSURE DEPENDENCE OF SOME BORON-RICH SOLIDS. J.S.Loveday, R.J.Nelmes, W.G.Marshall, Department of Physics and Astronomy, The University of Edinburgh, Edinburgh, UNITED KINGDOM.; J.M.Besson, S.Klotz, Physique des Milieux Condenés (CNRS), Université P. et M. Curie, Paris, FRANCE; G.Hamel, Département des Hautes Pressions, Université P. et M. Curie, Paris FRANCE.

The boron-rich solids (BRS) are a group of materials whose unusual transport properties and refractory nature offer potential applications in high-temperature electronic devices. The unusual icosahedral structures of the BRS, and the related three-centre electron deficient bonding, are thought to play a role in the transport properties via inverted-molecular (IM) compression - the icosahedral structural units are more compressible than the structure linking them together. High-pressure diffraction studies can provide direct tests of assumptions about bonding strengths in the BRS. Our neutron-diffraction studies of the boron carbide B4C have already given the first direct evidence of IM compression, and a secure structural basis for models of the electrical conductivity. The results showed that the behaviour is more complex than previously thought. Now, some recent measurements of the effect of changing carbon composition in boron carbides, and studies of the structurally related B6O and B6P, give results that differ significantly from currently accepted models of the bonding and bond strengths. These new results will be presented.

DD3.11 
PRESSURE STRUCTURAL PHASE TRANSITION IN Ti, Zr, AND Hf. Gerald Jomard, Alain Pasturel, LPMMC, CNRS, Grenoble, FRANCE; Laurence Magaud, LEPES, CNRS, Grenoble, FRANCE.

On the basis of first-principles total energy calculations, we have studied the crystal structures of Ti, Zr and Hf under pressure. The three metals are shown to exhibit a crystal structure sequence hcp with increasing pressure in good agreement with experiment. A subsequent bcc transition is also found for the three metals, which is also in agreement with experiment for Zr and Hf whereas the bcc structure for Ti is a prediction. We show that a full-potential linear-muffin-tin-orbitals (FP-LMTO) method coupled with the generalized gradient approximation (PW 91) of Perdew and Wang are found to be necessary to obtain calculated transition volumes as well as transition pressures in fairly agreement with experiment. The chemical bonding of the structure is shown to be quite different from what it is normally the case in transition metals, with a large degree of covalency. Despite the difference in the character of the chemical bonds for the three structures, we can to some extent use band-filling arguments to explain the crystal sequence of these metals.

DD3.12 
EQUATION OF STATE AND PHASE TRANSFORMATION STUDIES OF BeS TO 96 GPa. Chandrabhas Narayana, Victor J. Nesamony and Arthur L. Ruoff, Cornell University, Department of Materials Science and Engineering, Ithaca, NY; Anna University, Department of Physics, Madras, INDIA.

We report a reversible first-order phase transformation of the hygroscopic compound BeS studied in a diamond anvil cell using energy dispersive x-ray diffraction with a synchrotron source up to 96 GPa at ambient temperature. On loading, BeS transforms from the zincblende (ZB) structure, B3, to the nickel arsenide (NiAs) structure, B8, at around 59 GPa which is stable to 96 GPa. On unloading, at around 43 GPa BeS reverts back to ZB structure. The B3 to B8 transformation has a volume change of 11 % and the equilibrium transition pressure is 51 GPa. A second-order Birch equation discribes the equation of state of the B3 phase with bulk modulus Bo = 105 GPa and B. Theoretical calculations agree closely [1,2].

DD3.13 
X-RAY STRUCTURAL INVESTIGATION OF Li3N: FROM AMBIENT PRESSURE TO 35 GPa. Allen C. Ho, Maurice K. Granger, Arthur L. Ruoff, Cornell University, Department of Materials Science and Engineering, Ithaca, NY.

The crystal structure of the I-V compound Li3N has been studied in a diamond anvil cell using energy-dispersive x-ray diffraction with a synchrotron source over the pressure range of 0 to 35 GPa at ambient temperature. Prior to loading, both the hexagonal IrAl3 (P63/mmc) and the hexagonal AlB2 (P6/mmm) structures were present. Upon loading, the AlB2 phase was completely transformed to the IrAl3 phase below 8 GPa. The equation of state of the IrAl3 phase was fitted with a second-order Birch equation and the bulk modulus and its pressure derivative were determined. Using this equation of state, theortical calculations of lithium nitride's chemical potential have been made to 35 GPa at ambient temperature.

DD3.14 
PRESSURE-TUNING OF INTERMEDIATE VALENCE THERMOELECTRIC MATERIALS. Deborah A. Polvani, John V. Badding, Masashi Hasegawa, The Pennsylvania State University, Dept of Chemistry, University Park, PA.

The development of improved thermolectric materials would lead to important advances in technologies such as refrigeration, electric power generation, and cooling of electronic and superconducting components. Conventional searches for improved thermoelectric materials included synthesizing a large number of compounds and then investigating their thermoelectric properties as a function of composition or doping levels. We synthesize rare-earth transition metal pnictides and pressure-tune their interaction parameters with a diamond anvil cell to greatly reduce the large number of materials that need to be prepared. Pressure-tuning each of these samples will provide fundamental insight into the parameters necessary for high thermopowers and should suggest means to reproduce the high thermopowers at ambient pressures.

DD3.15 
HIGH-PRESSURE SYNTHESIS OF POTASSIUM-TRANSITION METAL COMPOUNDS. Charles M. Leland, Trent S. Snider, Laura J. Parker, and John V. Badding, Pennsylvania State Univ, Dept of Chemistry, University Park, PA.

Under pressure, the heavy alkali metals (potassium, rubidium, and cesium) undergo an s to d electronic transition, adopting a d1 electron configuration. The transition metal-like character of these metals then allows reactions between alkali metals and transition metals that do not normally occur at ambient pressure. Such phases may exhibit interesting properties due to their novel electronic character. In recent publications, we have shown such reactions between potassium and the transition metals nickel, palladium, and silver. Newer systems of interest include potassium-ruthenium, potassium-rhodium, and potassium-copper, which will be discussed in this work.

DD3.16 
HIGH-PRESSURE SYNTHESIS AND MAGNETIC PROPERTIES OF CHROMIUM DIANTIMONIDE. Hirotsugu Takizawa, Kyota Uheda, Tadashi Endo, Tohoku Univ, Dept of Materials Chemistry, Sendai, JAPAN; Masahiko Shimada, Tohoku Univ, Inst for Advanced Materials Processing, Sendai, JAPAN.

A new ferromagnetic polymorph of CrSb2 was synthesized under high-pressure/ temperature conditions of 7 GPa and 600-650C using the Belt-type high-pressure equipment. The crystal structure is body-centered tetragonal with the space group I4/mcm, which is refereed as CuAl2-type structure. The pressure-temperature formation diagram revealed that the high-pressure phase was formed above 5.5 GPa, and the compound crystallized into the marcasite-type structure below 5 GPa. The characteristic of the high-pressure phase is the metallic bond nature including the formation of Cr-Cr-Cr linear chain along the c-axis. This is in contrast with the fact that each chromium atom ionically bonds to Sb-Sb diatomic anions to form edge-shared octahedra in low-pressure marcasite-type phase. Although the low-pressure polymorph is an antiferromagnetic semiconductor with localized d2 electron configuration, the high-pressure form shows metallic conductivity and itinerant electron ferromagnetic behavior with the Curie temperature of ca. 160 K.

DD3.17 
STRAIN DISTRIBUTION IN NON-HYDROSTATICALLY COMPRESSED AU, MO, AND RE TO 42 GPA. Thomas Duffy, Princeton Univ, Dept of Geosciences, Princeton, NJ; Guoyin Shen, Univ of Chicago, CARS, Chicago IL; Ho-kwang Mao, Russell Hemley, Geophysical Lab, Washington DC; Dion Heinz, Univ of Chicago, Dept of Geophysical Sciences, Chicago IL.

The stress state in all static high-pressure experiments above 10 GPa is non-hydrostatic to some degree. Detailed characterization of this stress state is essential for determining material strength and for proper interpretation of high-pressure equation of state measurements. Due to the restrictive geometry of the diamond cell, the usual powder diffraction experiment provides measurements of strain only near the minimum stress direction. We have used a side diffraction geometry with x-ray transparent gaskets to characterize the complete variation of strain with angle from the diamond cell stress axis. Experiments were carried out at 13-BM-A of the GSECARS sector of the Advanced Photon Source. A two-circle diffractometer and energy-dispersive diffraction was used. Two samples were studied: a mixture of gold and rhenium was examined to 42 GPa, and a sample of gold and molybdenum was studied to 25 GPa. A large strain anisotropy was observed. The apparent pressure calculated from molybdenum diffraction lines was found to vary from 12 GPa to 24 GPa as the diffraction direction was varied from the minimum to maximum stress direction. Using the theory for deviatoric strain in a non-hydrostatically compressed sample, the shear strength and elastic moduli of Au, Re, and Mo were determined as a function of pressure.

DD3.18 
SINGLE CRYSTAL GROWTH OF LEAD TITANATE VIA HYDROTHERMAL SYNTHESIS AT SEVERE CONDITIONS. M. C. Gelabert, R.E. Riman, Department of Ceramic Engineering, Rutgers University, Piscataway, NJ; R.A. Laudise, Lucent Technologies, Murray Hill, NJ; L. E. McCandlish, Ceramare Corporation, Highland Park, NJ.

We report the growth of lead titanate under supercritical hydrothermal conditions (T > 350 C, P 30,000 psi) as part of a study on growing large single crystals of ferroelectric materials under a wide range of hydrothermal conditions. Syntheses were performed in welded platinum capsules in autoclaves at about 500 C and 30 kpsi. Growth of 100-m PbTiO3 crystals is presented.

8:30 AM *DD4.1 
HIGH PRESSURE LATTICE INSTABILITIES AND STRUCTURAL PHASE TRANSITIONS IN SOLIDS FROM AB-INITIO LATTICE DYNAMICS. Stefano Baroni, SISSA and INFM, Trieste, ITALY, and CECAM, Lyon, FRANCE.

Solids under an applied pressure display a variety of mechanical instabilities which may determine a transition to a structural phase other (usually of lower symmetry) than the ambient-pressure equilibrium one. These instabilities may result from the softening of some sound velocity (shear instability), from the softening of some phonon mode, or, as it often happens, from a combination of the two. In this talk I will discuss how ab-initio lattice-dynamical calculations based on density-functional perturbation theory provide powerful guidelines for predicting the existence of these lattice instabilities and for understanding the nature of the resulting low-symmetry phases. In this context, I will also present a new method for determining the sound velocities of complex crystal structures, along with its application to some problems of geophysical interest. As an illustration of this general methodological framework, I will report on recent studies on the high-pressure phases of cesium halides and hydride, and on the relations between lattice instabilities and the pressure-induced amorphization of quartz and related materials.

9:00 AM *DD4.2 
NEW PHASE AND AMORPHIZATION OF SILICA UNDER PRESSURE. James R. Chelikowsky, Department of Chemical Enginnering and Materials Science, Minnesota Supercomputer Institute, University of Minnesota, Minneapolis, MN.

Using first principles variable cell shape molecular dynamics simulations, we have found a new structure for silica. This structure results from annealing quartz under pressure near a strong phonon instability in the lowest acoustic branch. The new phase is obtained by rotations of SiO4 tetrahedra producing highly distorted edge-sharing octahedra and four-, or nearly five-fold, coordinated silicon polyhedra. This is the type of mechanism which Stolper & Ahrens (Geophys. Res. Lett. 14, 1231 (1988)) suggested would produce coordination changes in amorphous and liquid silicates under pressure. Here it is observed in a crystalline to crystalline phase transformation for the first time. The diffraction pattern of the new phase compares favorably with that of the unidentified intermediate crystalline phase found in quartz before it amorphizes (Kingma et al, Phys Rev. Lett. 70, 3927 (1993)). This suggests that pressure induced amorphization, can, in certain cases, be produced by widespread lattice instabilities. The new phase is found to be racemic and we anticipate it should display chiral properties.

9:30 AM *DD4.3 
FREE ENERGY DIFFERENCE CALCULATIONS OF PHASE TRANSITIONS: LATTICE SWITCH MONTE CARLO. G.J.Ackland, N.B.Wilding and A.D.Bruce, Department of Physics and Astronomy, University of Edinburgh, Edinburgh, SCOTLAND.

A new method of calculating free energy differences between phases is presented. It involves Monte Carlo sampling of the displacements of the atoms from lattice sites, and an additional lattice switch move which translates the lattice sites from one phase to another. In order that the switching occurs reasonably frequently, it is necessary to weight the sampling distribution in favor of those states which allow the switch. The method provides a direct measure of free energy differences, without the need for thermodynamic integration or definition of a reference state. An application to the hcp-fcc free energy difference for hard spheres is presented, and applications to more realistic systems are discussed.

10:00 AM DD4.4 
LASER-DRIVEN EQUATION OF STATE DATA FOR LOW-Z MATERIALS FROM 0.2 TO 40 MBAR*. R. Cauble, L. B. Da Silva, T. S. Perry, D. R. Bach, K. S. Budil, P. Celliers, G. W. Collins, D. M. Gold, J. Dunn, B. A. Hammel, N. C. Holmes, J. D. Kilkenny, R. E. Stewart, and R. J. Wallace, Lawrence Livermore National Laboratory, Livermore, CA.; and A. Ng, Dept of Physics, University of British Columbia, Vancouver, CANADA.

High intensity lasers offer the opportunity to explore the equations of state (EOS) of materials under conditions more extreme than any other laboratory method. However there have been few laser-driven EOS experiments because of the difficulties in the technique. We have used both direct laser irradiation and indirect laser-generated x rays to drive strong shocks and perform EOS measurements on several low-Z materials. Employing side-on radiography, we have obtained data on the principal Hugoniots of liquid deuterium from 0.25 to 2 Mbar (25-200 GPa), polystyrene (CH) from 10-40 Mbar (1-4 TPa), and beryllium near 15 Mbar. The deuterium and CH results were surprising. The Be data made it possible to confirm a large set of previous EOS work involving materials shocked as high as 65 Mbar.

10:30 AM *DD4.5 
THE RUBY SCALE AT MAGABAR PRESSURES. Isaac F. Silvera, Harvard University, Physics Department, Cambridge, MA.

Diamond anvil cells revolutionized ultra high pressure research because of the gasketed sample and the ruby pressure scale. The ruby scale provided an easy accurate method of determining pressure distributions within samples. However, at pressures above 100 GPa measurements of the ruby fluorescence begin to become difficult due to weakening signal, interference from intense diamond fluorescence, and ultimately, strong absorption of the exciting laser by diamond itself. I shall discuss a number of methods which have been devised to extend the ruby pressure calibration method to 500 GPa, including quasi-direct pumping of the ruby R-lines.

11:00 AM DD4.6 
ULTRAHIGH PRESSURE DIAMOND CELL WITH A BERYLLIUM SIDE WINDOW. Ho-kwang Mao and Russell J. Hemley, Geophysical Laboratory and Center for High Pressure Research, Carnegie Institution of Washington, Washington, DC.

A new type of megabar high-pressure cell with opposing diamond anvils and beryllium gasket has been developed to free several technical limitations of conventional diamond cells. Beryllium provides a side window for x-ray studies in the radial direction which is inaccessible previously. With the new x-ray diffraction geometry, the single-crystal elasticity tensor, shear strength, and preferred orientation of hcp iron have been obtained up to a pressure of 220 GPa. The low attenuation of beryllium also allows high-pressure x-ray spectroscopic study to be extended from 12 keV down to 5 keV. Direct characterizations of electronic and magnetic properties of materials at ultrahigh pressures have become feasible.

11:15 AM DD4.7 
STUDIES ON MATERIALS UNDER ULTRAHIGH P-T AT THE ADVANCED PHOTON SCOURCE. Guoyin Shen, Thomas Duffy*, Yanbin Wang, Mark Rivers, and Stephen Sutton, Consortium for Advanced Radiation Sources, The Univ of Chicago, Chicago, IL; *Now at: Dept of Geosciences, Princeton Univ, Princeton, NJ.

The Advanced Photon Source (APS) is a third-generation synchrotron now in operation at Argonne National Laboratory. A national user facility for research in the Earth sciences is being constructed by the GeoSoilEnviro group of the Consortium for Advanced Radiation Sources (GSECARS). One component of this effort is a program for diamond anvil cell (DAC) research. The scientific goal of the DAC program is to study geological materials across the entire pressure-temperature spectrum of the terrestrial planets by a variety of high pressure experiments at ambient and high temperature. While ultrahigh pressures and temperatures are generated in a diamond anvil cell, the achievement is at the expense of reducing sample volume and there usually exist large P-T gradients across a sample. To improve the accuracy of experimentally determined quantities, efforts are being made to have characterization capabilities with high spatial resolution and to reduce the P-T gradients in sampling area. High brilliance synchrotron radiation sources such as the APS are, therefore, ideal for ultrafine and sensitive X-ray measurements. A double sided laser heating system allows us to obtain uniform temperatures in a high pressure sample. The combined laser heating/x-ray energy dispersive diffraction system has been used to obtain in situ structural data on metals, alloys, and silicates to temperatures of 4000 K and pressures above 90 GPa. The first experiments were conducted in December, 1996 in collaboration with scientists from the Carnegie Institution of Washington and from the University of Chicago. The experiments used a two-circle diffractometer and a side diffraction geometry to measure the dependence of lattice parameter on angle from the diamond cell stress axis in a sample. These experiments constrain the shear strength and shear modulus of these materials at high pressure. Experiments on variety of materials will be reported together with results on high pressure melting on iron and its high P-T phase transitions.

11:30 AM DD4.8 
A MULTI-ANVIL, HIGH PRESSURE SYSTEM WITH SYNCHROTRON X-RAY PROBES: NEW OPPORTUNITIES FOR IN-SITU MATERIALS RESEARCH AT HIGH PRESSURE AND TEMPERATURE. Yanbin Wang, Guoyin Shen, Mark Rivers, Steve Sutton, and Peter Eng, Consortium for Advanced Radiation Sources, The University of Chicago, Chicago, IL.

A multi-anvil, high-pressure facility is being constructed at GeoSoilEnviroCARS (Center for Advanced Radiation Sources, the University of Chicago), Sector 13 at the Advanced Photon Source, Argonne National Laboratory. The sector consists of an undulator (13-ID) and a bending magnet beamline (13-BM), with two multi-anvil systems (MAS) installed at the end stations 13-BM-D and 13-ID-D. A 250 ton MAS system is currently being installed at 13-BM-D and expected to be operational in October, 1997. A 1000 ton system will be operational at 13-ID-D early 1998. These MAS are capable of generating simultaneously high pressures and temperatures up to 40 GPa and 3000ƒC, with millimeter sized samples at the high-pressure end and centimeter sized samples at low pressure end. Resistance heating provides a homogeneous temperarure distribution that is stable for hours or days under pressure. Such high temperature minimizes unwanted non-hydrostatic stresses and pressure gradient and is essential in making high-quality measurements. By optimizing the system, the pressure and temperature range of interest can be swept through with one single experiment, with either energy-dispersive or monochromatic X-ray diffraction options. The large sample volumes offered by these MAS allow endless modifications in the pressure cell, making it possible to carry out an extremely wide range of experiments with a variety of in-situ characterization techniques. Currently, our MAS facility is aiming at the following areas of research: (1) High-resolution monochromatic powder X-ray diffraction to discover new phases and study crystal chemistry under high pressure and temperature; (2) Phase equilibrium studies with in-situ pressure and temperature determination, (3) Time resolved experiments with fast data collection to subseconds, to study kinetics of phase transformations. Monochromatic powder diffraction can be employed to understand transition mechanisms. (4) A modification in the pressure tooling allows deformation experiments with controlled strain or strain rate, allowing rheological properties of solids to be measured under high pressure and temperature; (5) In addition to the conventional P-V-T equation-of-state measurements, simultaneous measurements of density and acoustic velocities will be carried out using ultrasonic techniques coupled with X-ray diffraction. This will bring a new level of pressure calibration and allow an equation-of-state determination without relying on any previously determined pressure standard.

11:45 AM DD4.9 
SOUND VELOCITY MEASUREMENTS IN OXIDES AND SILICATES AT SIMULTANEOUS HIGH PRESSURES AND TEMPERATURES USING ULTRASONIC TECHNIQUES IN MULTI-ANVIL APPARATUS IN CONJUNCTION WITH SYNCHROTRON X-RADIATION DETERMINATION OF THE EQUATION OF STATE. Robert C. Liebermann, Ganglin Chen*, Baosheng Li*, Gabriel D. Gwanmesia**, Jiuhua Chen*, Michael T. Vaughan*, Donald J. Weidner; Center for High Pressure Research and Department of Earth and Space Sciences, SUNY, Stony Brook, NY; *Center for High Pressure Research and Mineral Physics Institute, SUNY, Stony Brook, NY; **Department of Physics and Astronomy, Delaware State University, Dover, DE.

Li et al. [PEPI, 98, 79-91, 1996] have developed specimen friendly cell assemblies for multi-anvil, high-pressure apparatus which enable precise ultrasonic interferometeric measurements of wave velocities in minerals to be performed to P>12 GPa and T>1300 K. These experiments initially utilized a 1000-ton Kennedy-Getting type press and a split cylinder module of Walker's design (USCA-1000) combined with an ANUTECH ultrasonic interferometer to study both polycrystalline and single crystal specimens. We have recently adapted these techniques for use in a DIA-type, cubic-anvil apparatus (SAM 85) installed on the superconducting wiggler beamline (X17B) at the National Synchrotron Light Source of the Brookhaven National Laboratory. X-ray spectra of both the polycrystalline specimen and the NaCl medium which surrounds it are monitored continuously; the former provides PVT data to compliment the velocity measurements and the latter the pressure standard. Studies on Lucalox alumina to 9.7 GPa showed behavior very consistent with the data from the experiments in the USCA-1000 apparatus by Li et al., in which pressure was estimated from observing the phase transitions in Bi and ZnTe. Subsequent experiments achieved simultaneous conditions of 9 GPa and approximately 1500 K. Experiments on single-crystal MgO to simultaneous pressures of 9 GPa and temperatures of 1500 K are compared to the acoustic measurements of Spetzler to 0.8 GPa and 800 K, and of Jackson and Niesler to 3 GPa and Yoneda to 6 GPa at 298 K. There is excellent consistency in the pressure derivatives of the elastic moduli determined in these different studies. The large P-T range and precise determination of the sample cell pressures allow us to extract the mixed derivatives of the elastic moduli of MgO unambiguously, an objective which has heretofore eluded experimentalists. The mixed P-T derivatives for the two compressional modes (c11 and c110) are of the order of 10-3/K while that for the shear mode (c44) is at least an order of magnitude smaller. New data on P and S velocities in the polycrystalline MgSiO3-majorite to 7 GPa at 1000 K yield pressure derivatives of the bulk and shear moduli that are similar to those of pyrope (Mg3Al2Si3O12), but somewhat lower that the values obtained by Rigden et al. for a Py62Mj38 specimen. Finally, we report new data on P and S velocities in the polycrystalline olivine and wadsleyite phases of Mg2SiO4 to 7 GPa and 900 K and compare these new data to the acoustic measurements of Yoneda and Morioka to 8 GPa, Duffy et al. and Zha et al. to 16 GPa, and Li et al. to 13 GPa, all at 298 K, and with other equation of state data. These new developments open new opportunities for establishing absolute pressure scales and for making ultrasonic measurements at simultaneous high pressures and temperatures in conjunction with X-ray determinations of the cell pressure and sample volume.

12:00 NOON DD4.10
AND NUCLEAR SPIN RELAXATION AT HIGH PRESSURE IN MEOH AND 4:1 MEOH-ETOH MIXTURES. R.F. Marzke, Arizona State University, Dept of Physics and Astronomy, G.H. Wolf, R. Nieman, Dept of Chemistry and Biochemistry, Tempe, AZ; J. Yarger, University of California, Department of Chemistry, Berkeley, CA; D.P. Raffaelle, Glendale College, Glendale, AZ.

Diffusivity and Nuclear Spin Relaxation times T2, T1have been measured to 4.0 GPa for both pure MeOH and 4:1 MeOH-EtOH mixtures, using NMR in the diamond anvil cell. In pure MeOH, D-1and T2show identical activated pressure behavior, with an activation volume significantly different from that of viscosity. This implies a pressure-dependent infinite-frequency shear modulus. The Stokes-Einstein relation is not valid with a constant hydrodynamic radius,to pressures above 1 GPa, but is modified to take account of the Maxwell relation between viscosity and correlation time. This leads to indirect measurement of the shear modulus, via viscosity and NMR measurements, over a wide pressure range.

SESSION DD5: HIGH-PRESSURE SYNTHESIS AND SUPERHARD MATERIALS 
Chair: Paul Alivisatos 
Tuesday Afternoon, December 2, 1997 
Staffordshire (W)

1:30 PM *DD5.1 
MATERIALS RESEARCH AT THE INSTITUTE FOR HIGH PRESSURE PHYSICS OF THE RUSSIAN ACADEMY OF SCIENCES. Sergei M. Stishov, Institute for High Pressure Physics of the Russian Academy of Sciences, Troitsk, Moscow Region, RUSSIA.

The L.F. Vereschagin Institute for High Pressure Physics of the Russian Academy of Sciences (IHPP) is a unique scientific organization which is devoted almost completely to theoretical, experimental, and applied aspects of compressed matter. The Institute, founded by L.F. Vereschagin in 1958, was internationally recognized in the 1960's because of the successful synthesis of diamond and cubic boron nitride. The original equipment and technology developed at the IHPP were a basis for industrial production in the former USSR. The subsequent synthesis of dense silica with the rutile structure strongly influenced the direction of high pressure studies internationally in geophysics and planetary physics. This work finally established the Institute as a first rate scientific organization. Despite the significant loss of staff and meager funding in recent years, IHPP continues to perform research in materials science, which I will describe briefly: (1) Phase transitions in solids and liquids, metastable phases, and amorphous substances and glasses at high pressure. A number of new results will be reported, including softening of the sheer modulus of ice prior to amorphization and the amorphization of fullerites at high pressures and temperatures. (2) Growth of single crystals of high pressure phases and their properties, including coesite, compounds of magnesium with the elements of the fourth group, and boron nitride. X-ray analysis reveals very complicated crystal structures for the Mg-compounds. (3) Compressibilities of disordered systems. Macroscopic compressibilities will be compared with microscopic ones obtained from X-ray and EXAFS stuies. Data on logarithmic volume relaxation in glasses and powders will be presented. (4) Superhard materials and tool fabrication, including diamond, polycrystalline composites of boron nitride, and compacted materials and their fabrication into tools. (5) Geophysical studies and fabrication of carbon-carbon composite materials with the world's largest laboratory press.

2:00 PM *DD5.2 
HIGH PRESSURE SYNTHESIS AND CHARACTERIZATION OF SOLID STATE MATERIALS. J. V. Badding, Dept. of Chemistry, Pennsylvania State University, University Park, PA.

This talk will present the synthesis and characterization of solid state materials at high pressures from the perspective of a solid state chemist. Efforts in the following areas will be described: 1) Synthesis of novel carbon-based networks and polymers 2) Synthesis of novel alkali metal-transition metal compounds and 3) Pressure tuning and synthesis of intermediate valence rare-earth based thermoelectric materials.

2:30 PM *DD5.3 
CARBON ONIONS AS NANOSCOPIC PRESSURE CELLS FOR DIAMOND FORMATION: A FIRST PRINCIPLES STUDY. Alessandro De Vita, IRRMA, J.-C.Charlier, Université Catholique de Louvain; X. Blase, Université Claude Bernard-Lyon I; F. Banhar, Max-Planck-Institut für Metallforschung; P.M. Ajayan, Rensselaer Polytechnic Institute; and R.Car, IRRMA

In recent experiments, Banhart and Ajayan have shown that graphitic particles with a concentric polyhedral structure (``carbon onions'') transform to diamond if heated to 700 0C and irradiated with electrons. Diamond nucleates after the onion-like particles ``shrink'' to a non-uniform interlayer spacing pattern, decreasing from 3.3 Å at the outside of the onion to values as low as 2.2Å at the centre. In the present work we study the microscopic mechanisms that lead to the observed shrinking of the shells. The role played by defective ring topologies such as pentagon-octagon-pentagon and heptagon-pentagon units along a proposed pathway of shell surface reduction will be discussed. Results from first principle simulations will be used to discuss the role of interstitial defects and inter-shell chemical bonds in the reduction process. Finally, we will show how all these mechanisms can be recast into an elastic model which reproduces the observed interlayer spacing.

3:15 PM *DD5.4 
HIGH PRESSURE - HIGH TEMPERATURE SYNTHESIS OF LOW COMPRESSIBILITY CUBIC C3N4. Jeffrey H. Nguyen, University of California, Department of Physics and Department of Geology and Geophysics, Berkeley, CA, now at Lawrence Livermore National Laboratory, Physics and Space Technology, Livermore, CA; Wendel A. Caldwell, University of California, Department of Geology and Geophysics, Berkeley, CA; Laura Robin Benedetti, University of California, Department of Physics, Berkeley, CA; Michael B. Kruger, University of Missouri, Department of Physics, Kansas City, MO; Raymond Jeanloz, University of California, Department of Geology and Geophysics, Berkeley, CA.

We report the synthesis of a dense carbon-nitride in bulk form at high pressures and high temperatures. Synchrotron x-ray diffraction patterns of the new phase are compatible with that of the cubic C3N4 structure predicted by Teter 1. In order to determine optimal synthesis conditions for this material, an array of experiments was carried out over a broad range of pressures (7-60 GPa) and temperatures (1000-5000 K) using the laser-heated diamond cell and a variety of starting materials, including carbon + nitrogen. The equation of state obtained from the high pressure x-ray diffraction patterns reveals that this phase is relatively incompressible.

3:45 PM DD5.5 
HIGH-PRESSURE CHEMISTRY OF CARBON NITRIDE MATERIALS. Andrew J. Stevens, Carl B. Agee, Charles M. Lieber, Harvard University, Departments of Chemistry and Earth and Planetary Sciences, Cambridge, MA.

The composition, structure and properties of the carbon nitride sp2-bonded precursor paracyanogen has been studied at high pressures and temperatures. At high-pressures, paracyanogen decomposes to carbon and molecular nitrogen with the decomposition temperature increasing with pressure over the range of 5 to 20 GPa. Prior to decomposition paracyangoen can be transformed to a atmospheric pressure quenchable phase that is 25% higher in density and over an order of magnetude harder than the starting paracyanogan material. Structural analyses of this quenchable phase show, however, that it consists of a sp2-bonded carbon-nitrogen network. In addition, the decomposition kinetics of the paracyanogen have been studied in detail. At 15 GPa, the initial decomposition exhibits a large prefactor and an activation energy of 2.8 eV. The decomposition kinetics also have been found to change when the average N-N separation exceeds a critical distance. The implications of these results to the high-pressure synthesis of a superhard, sp3-bonded carbon nitride solid will be discussed.

4:00 PM DD5.6 
HIGH PRESSURE SYNTHESIS OF THE NEW SUPERHARD SUBSTANCES. Vadim Brazhkin, Alexander Lyapin, Institute for High Pressure Physics RAS, Troitsk, Moscow, RUSSIA.

Short review of the attempts to prepare new superhard compounds under high pressure is presented. There are two main types of superhard materials: i) metastable high-pressure forms of the ordinary substances (C, BN, SiO2 etc); ii) new compounds, which have not counterparts at normal pressure (C3N4). We analyze the relation between mechanical characteristics (hardness, strength) and elastic moduli for the hard materials. The problem of comparison of the hardness for different superhard substances are also considered. The "ideal" hardness of each material tends to 0.5E, where E is the Young modulus. The bulk modulus B has not any rigorous correlation with hardness. The new class of superhard metastable carbon materials prepared from fullerites C60 have been recently discovered. We have studied high pressure-high temperature transformations in C60 and C70 crystals and in a mixture of different fullerites C2n (50<2n<170). The two kinds of superhard carbon phases were prepared: (i) three-dimensionally (3D) polymerized fullerites and (ii) amorphous sp2-sp3 carbon phases. The polymerized fullerites have the Young modulus near 400 GPa (by 3 times less than that of diamond) and hardness in the range 20-60 GPa (by 1.5-5 times less than that of diamond). Amorphous carbon phases with large share of sp3 cites (40-80 ) have Young moduli up to 700 GPa and hardness in the range 20-90 GPa, depending on the density and share of sp3 cites. The amorphous carbon materials have large values of fracture toughness coefficient (7-15 MN/m-3/2), higher than that of diamond. 3D polymerized fullerites, being very hard, can be even plastic at room temperature. The prospects for the search of new pressure-synthesized superhard materials are discussed.

4:15 PM DD5.7 
HIGH-PRESSURE, HIGH-TEMPERATURE SYNTHESIS OF SUPER-HARD BORON-RICH SOLIDS. Herve Hubert, Materials Research Center, Dept. of Chemistry & Biochemistry, Arizona State University, Tempe, AZ; Bertrand Devouard, Laurence A.J. Garvie, Peter R. Buseck, Depts. of Geology and Chemistry & Biochemistry, Arizona State University, Tempe, AZ; Michael O'Keeffe, William T. Petuskey, and Paul F. McMillan, Materials Research Center, Dept. of Chemistry & Biochemistry, Arizona State University, Tempe, AZ.

Boron-rich solids, with a structure related to that of -rhombohedral boron, are refractory compounds which combine low-density, high-hardness, and chemical inertness. They also display potentially useful thermal and semi-conducting properties for thermoelectric power generation. This family of materials includes boron carbides and boron suboxide (nominally B6O), the third hardest known material. In this high-pressure, high-temperature investigation, we synthesized -rhombohedral phases belonging to the B-C-N-O system using a multianvil press. The B-C-O compounds were obtained by reducing B2O3 with B or mixtures of B and C between 1 to 10 GPa at temperatures between 1200 and 1800C. The samples were characterized by powder X-ray diffraction, scanning electron microscopy, transmission electron microscopy, parallel electron energy-loss spectroscopy (PEELS), and photoluminescence. Samples of B6O of improved purity and crystallinity (in comparison to room-pressure syntheses) were obtained at high pressure. Quantitative analysis, performed using PEELS, showed the samples to be less oxygen-deficient (i.e., closer to the nominal B6O composition) than other reported materials, with compositions ranging from B6O0.86 to B6O0.96. We obtained euhedral particles from 20 nm to 30 m, the larger being icosahedral twins. We showed that B6O closely matches the geometrical requirements for icosahedral twinning.We studied the influence of pressure on the cell parameters, grain morphology, defects types and density, stoichiometry, and electronic properties. We also synthesized several intermediate phases between B6O and B4C, with compositions from B4C0.73O0.22 to B4C0.86O0.2. Crystals up to 20 m were observed. In this case, the cell dimensions preclude the formation of icosahedral twins, but some 5-fold cyclic twins are encountered. We also report the first bulk synthesis of B6N1-x, nitride anologue of B6O, obtained by reacting B and BN at 7.5 GPa and 1700C.

4:30 PM DD5.8 
LASER HEATING OF BORON NITRIDE IN DIAMOND ANVIL CELL. Mikhail Eremets, NIRIM, Tsukuba and CREST, JST JAPAN, ISAN and CSM, Troitsk, RUSSIA, Kenichi Takemura, Hitoshi Yusa, Dmitri Golberg, Yoshio Bando, Yoichiro Sato, Kenji Watanabe, NIRIM, Tsukuba, JAPAN, Vladimir Blank, ISAN and CSM, Troitsk, RUSSIA.

Platelets of single-crystalline cubic BN or polycrystalline hexagonal BN in a diamond anvil cell were directly heated through absorption of CO2-laser radiation by optical phonons at pressures up to 15 GPa. The recovered samples were analyzed by high resolution and scanning electron microscopy, micro-Raman, and X-ray diffraction using synchrotron radiation. Different phenomena were observed in the result of the heating. In particular, a microscopic mechanism of the diffusion-like solid-solid first-order transitions was studied. BN exhibits an extreme case of transitions between very different lattices of hBN and cBN phases, accompanied with the large change of the volume. A boundary between the hexagonal and cubic phases was produced by the laser local heating. In the transient region between the phases a mixture of highly fragmented disordered nanocrystallites and amorphous state of BN was found. Analysis of the pressure dependence of the phase transformations showed that the hBN-cBN-liquid triple point is located at 9(I1(J1 GPa in the pressure scale. Above this pressure melting of cubic BN was achieved for the first time. After the laser heating a polycrystal grow from the melt and a powder of microcystals from the vapor. hBN crystals are formed at pressures below 8 GPa, cBN ones above 10 GPa, and their mixture in the 8-10 GPa region. The results are discussed with respect to the phase diagram of BN and to the cBN film growth where high compressive stresses are believed to be important for nucleation. In the course of this study BN nanotubes were found in quite different cases: in the boundary between the cubic and hexagonal phases (both in the starting cBN and hBN samples) and at the surface of the both hBN and cBN microcrystals grown from the vapor in the 8-10 GPa region. In most cases the nanotubes grow from an disordered amorphous state at the surface of the crystals.

4:45 PMDD5.9 
ACCELERATED GENERATION OF c-BN, DIAMOND AND BCN UNDER HIGH PRESSURE AT HIGH TEMPERATURE DUE TO PREMILLING STARTING POWDERS. J.Y.Huang*, T.Taniguchi and S.Horiuchi, National Institute for Research in Inorganic Materials, Tsukuba, Ibaraki, JAPAN; *on leave from Laboratory of Atomic Imaging of Solids, Institute of Metal Research, Chinese Academy of Sciences, Shenyang, P.R. CHINA.

We have found that ball-milling of the starting h-BN powders can facilitate the phase transformation from h-BN to c-BN under high pressure at high temperature and increase c-BN transformation ratio tremendously. This is mainly due to the fact that ball milling is a heavy mechanical deformation process during which the materials are seriously deformed; according to high-resolution electron microscopy, different kinds of defects such as dislocations, twins and stacking-faults are induced into the materials. Dislocation analysis also showed that the introduction of defects in h-BN is favorable to c-BN generation. Based on these experimental result, extensive studies of the effect of pre-milling of graphite or graphite and h-BN mixture on the formation of diamond or BCN are in progress. The starting powders are deformed to different intensity, such as to very small grain size, partly or completely amorphous, and then pressed under high temperature and high pressure. A comparison with the non-milled specimens shows the effect of milling clearly. We can prospect an extensive industrial application of this new technique in the production of super-hard materials in the future since the property/cost ratio of c-BN, diamond and BCN can be decreased greatly.

5:00 PM DD5.10 
HIGH-PRESSURE RAMAN SCATTERING STUDY OF 4H AND 6H SiC TO 30 GPa. Murli H. Manghnani, V. Vijayakumar, Hawaii Institute of Geophysics and Planetology, School of Ocean and Earth Science & Technology, University of Hawaii, Honolulu, HI.

The linear positive dependences of Raman shift with pressure for the TO and LO modes in the high wavenumber region ( 770 - 974 cm-1) and corresponding mode-Gruneisen parameters, calculated from i = -(lni)/(lnV), are in good agreement with previous studies (e.g., Liu and Vohra, 1994). In contrast, the pressure dependences of the low-lying TA and LA modes in ( 100 - 270 cm-1 range), are mostly negative (especially for the 4H type), resulting in lower averaged values of the Gruneisen parameter ( 0.6). Implication of this is discussed in light of the different elastic properties and compressional behavior in the 6H and 4H polytypes.

SESSION DD6: HYDROGEN AT HIGH TEMPERATURE 
Chair: Russell J. Hemley 
Wednesday Morning, December 3, 1997 
Staffordshire (W)

8:30 AM *DD6.1 
DENSE HYDROGEN: STATES AND STATUS. N.W. Ashcroft, Cornell University, Ithaca, NY.

The fundamental Hamiltonian for a macroscopic quantity of hydrogen is simple, and it possesses considerable symmetry. Yet the emerging phase diagram is surprisingly complex at high density, but low temperature. This has been established by careful diamond-cell studies including structural, Raman, and infrared probing, the latter now indicating a possible state of spontaneous polarization. At higher temperatures the properties of dense hydrogen are becoming accessible through shock-wave studies which are yielding data on states which are both fluid and significantly conducting. In structural terms the nature of the fluid is unusual, and the physical mechanisms of conduction are also unusual.

9:00 AM *DD6.2 
FLUID HYDROGEN AT HIGH PRESSURES AND TEMPERATURES. N. C. Holmes and W. J. Nellis, Lawrence Livermore National Laboratory, Livermore, CA.

Hydrogen, the simplest and most abundant element in the universe, is proving to be among the most difficult to understand. A host of new shock compression experimental data have been developed in the past few years, posing new challenges to our understanding. I will describe new measurements of electrical conductivity and the discovery of metallic hydrogen, temperature measurements which imply molecular dissociation, and laser-driven shock experiments which test our understanding at very high pressures and densities. Further, we will discuss how the implications of these data and their interpretation have profound consequences on subjects as disparate as the interior of Jupiter, the quest for inertial confinement fusion, and space travel.

9:30 AM *DD6.3 
SOLID MOLECULAR PHASES OF HYDROGEN VIA CONSTANT-PRESSURE FIRST-PRINCIPLES MOLECULAR DYNAMICS. Jorge Kohanoff1, Sandro Scandolo1,2, Guido L. Chiarotti2 and Erio Tosatti1,2; 1ICTP, Trieste, ITALY; 2 SISSA, Trieste, ITALY.

By performing constant pressure ab initio molecular dynamics simulations we analyse the high pressure phases of molecular solid hydrogen. We use a gradient corrected LDA, and a freshly implemented efficient technique for Brillouin zone sampling. An extremely good k-point sampling turns out to be crucial for obtaining the correct ground state. Our constant pressure approach allow us to optimize simultaneously the orientational degrees of freedom, the lattice constants, and the space group. This can be done either by a local optimization technique, or by running molecular dynamics (MD) trajectories. The MD allows for the system to undergo structural transformations spontaneously. In the lower pressure regime, namely for the broken symmetry phase (BSP or phase II), we find a quadrupolar orthorhombic structure, of Pca21 symmetry. By means of an MD investigation, we find, at higher pressures, a slightly distorted orthorhombic structure very close to Cmca symmetry. This structure cannot be straightforwardly identified with the H-A phase (or phase III) because: 1) it is metallic, 2) the IR absorption is forbidden in the Cmca geometry, and 3) the Raman vibron discontinuity would be far too large compared to experiment. In fact, we argue that this phase is the first metallic molecular phase of hydrogen. Metallization would happen then, not via a band-overlap mechanism, but due to a structural transformation. By comparing total enthalpies, we also obtain an insight into the structure of phase III, which would still exhibit an insulating character.

10:30 AM *DD6.4 
NEW HIGH-PRESSURE EXCITATIONS IN PARA-HYDROGEN. A. F. Goncharov, R. J. Hemley, H. K. Mao, and J. F. Shu, Carnegie Institution of Washington, Washington, DC.

We have studied solid H2 by Raman and infrared spectroscopy to pressures in excess of 200 GPa and to 8 K. After cooling down to T<25 K we observe a ortho-para conversion in both phases I and II by monitoring the Raman spectra of the roton bands and the infrared spectra of the pure vibron and the zero-phonon rotation-vibration bands with time. We documented nearly complete ortho-para conversion for samples cooled at 76 and 97 GPa, while one, cooled at 110 GPa contained a substantial amount of ortho-species even after 112 hours at 15 K. For pure para-H2 the transition to the ordered phase (BSP, phase II) recorded at 110-120 GPa and 17 K is characterized by major and previously unobserved changes in rotation and rotation-vibration excitations. The second dramatic change of the low-frequency excitations (librons) occurs at 145-165 GPa, which is associated with the discontinuity in the Raman and infrared vibrons at the transformation to phase III. Librons in the phase III are strongly pressure dependent, indicating an increase in molecular ordering. New vibrons sidebands are observed in all three phases (I, II, and III) of p-H2.

11:00 AM *DD6.5 
QUANTUM AND CLASSICAL PHASE TRANSITIONS IN SYSTEM OF ROTORS. APPLICATION TO SOLID HYDROGEN AND OTHER SIMPLE MOLECULAR SOLIDS UNDER PRESSURE. Yuri Freiman, Serge Tretyak, Alexander Brodyanski, Verkin Institute for Low Temperature Physics, Kharkov, UKRAINE; Andrzej Jezowski, Trzebiatowski Institute for Low Temperature Physics and Structure Research, Wroclaw, POLAND.

Investigations of simple molecular cryocrystals (solid hydrogens, N2- and O2-type crystals) in the field of high and superhigh pressures revealed surprisingly rich phase diagrams; moreover, it was found that the phase diagram of each substance is essentially of individual nature. A large variety of phases with different positional and orientational structures and different types of orientational movement of molecules has been observed. In spite of this lack of generality, it is quite natural to look for general approaches to the nature of their phase diagrams, since in all the crystals under discussion the molecules interact via quasiquadrupolar intermolecular forces of the same type. We will discuss the nature of stability of different structures in terms of two parameters, U0 and U1, the molecular and crystal field constants, which are generated by, respectively, the coupling and single-molecular terms in the intermolecular interaction potential. It is shown that the orientational state of the system is determined by values and signs of the parameters U0 and U1 and can be described by the positive and negative order parameters. In the case of -N2 and the ordered phase of o-H2 U0 > 0, U1 = 0; for -O2 U0 < 0, U1 > 0 () and the order parameter is positive. For -O2 U0 < 0, U1 < 0 and the order parameter is negative and describes precession of disc-like molecules. The states with the negative order parameter can be treated as the orientational analog of the easy-plane-type ordering in magnets. The phase diagram of the system of rotors in coordinates U0, U1, T demonstrates a possible diversity of phases and types of phases transitions generated by the quadrupolar Hamiltonian. Along with the usual order-disorder transitions, order-order phase transitions are possible with or without change of the sign (and values) of the order parameter. Of particular interest is the range of the parameters characteristic of solid hydrogens under pressure. The phase diagram demonstrates the possibility of reentrant orientational phase transitions between a quantum crystal of hindered rotors and classical phase of librating quadrupoles. At positive molecular fields at certain positive values of the crystal field a critical point appears on the "classical" side of the phase separation line. With further grows of the crystal field another critical point appears at the "quantum side" of the phase separation line, that is the line turns into a segment which begins and ends with the critical points which eventually degenerates into a multicritical point and then disappears.

11:30 AM DD6.6 
PROGRESS IN THE SEARCH FOR SOLID METALLIC HYDROGEN. Arthur L. Ruoff and Chandrabhas Narayana, Cornell University, Department of Materials Science and Engineering, Ithaca, NY.

The experimental search for metallic hydrogen has been active for over two decades with the first claims being published in 1975, followed by several other claims all of which failed to pass the test of time. In 1989 scientific measurements were made on H2 at pressure up to about 210 GPa. This was followed by a hiatus until 1995 when our group showed that hydrogen was not yet an alkali metal at 290 GPa (measured at CHESS by the x-ray marker method based on tungsten.) In the present paper we describe Raman measurements on the vibron of H2 to the highest pressures attained. These show two important features: (1) The Raman shift vs. pressure decreases much less rapidly with pressure than expected from extrapolation of previous data at lower pressures. These results, combined with the theory of pairing instabilities developed by Ashcroft[1] suggest that the transition to an alkali metal should occur above 350 GPa rather than at 300 GPa. (2) The Raman peaks are strain broadened to such an extent that they may disappear before 350 GPa, so the vanishing of these peaks should not be taken as a sign of disappearing. In fact, hydrogen at ambient temperature is highly nonhydrostatic as evidenced not only by the Raman peak broadening but by the difference . The reason why it is so difficult to reach the required static pressure in solid hydrogen while 560 GPa was obtained on molybdenum [2] and why it may be difficult to prove metallization has occurred will be discussed.

11:45 AM DD6.7 
INSTABILITIES IN DENSE ATOMIC HYDROGEN. Andrew A. Quong, Sandia National Laboratories, Livermore, CA; Efthimios Kaxiras, Harvard University, Department of Physics, Cambridge, MA; Jeremy Q. Broughton, Naval Research Laboratory, Washington, DC.

Density-functional, linear response plane-wave calculations, supported by direct frozen phonon calculations indicate that, at densities typical of contemporary theoretical studies of atomic metallic hydrogen (i.e. rs = 1.31 au to 1.2 au), all of the prior reported candidate structures of hydrogen are unstable. In each case, phonons somewhere in the Brillouin zone have imaginary frequencies. The structures studied include face-center, body-center, simple and diamond cubic, beta-tin, and simple hexagonal in both minimum energy c/a ratio forms. The latter, in the filamentary low c/a form, the beta-tin and the diamond cubic are particularly relevant because these are the low coordination structures favored as the likely candidates for most stable atomic phase at the high pressure molecular to atomic transition. Our results imply that there may be no stable atomic crystalline phase and that upon applying pressure to the molecular phase, a transition to a fluid state occurs.

SESSION DD7: SEMICONDUCTORS AND SUPERCONDUCTORS AT HIGH PRESSURES 
Chair: Peter Y. Yu 
Wednesday Afternoon, December 3, 1997 
Staffordshire (W)

1:30 PM *DD7.1 
THERMODYNAMICS AND GROWTH OF GaN SINGLE CRYSTALS UNDER PRESSURE. S. Porowski, High Pressure Research Center Polish Academy of Sciences, Warsaw, POLAND.

GaN is recently considered as the most important material for blue and ultraviolet optoelectronics. The device structures are usually grown on foreign substrates which results in high density of dislocations above 108cm-3. The application of high N2 pressure gives a unique possibility of growing of GaN single crystalline substrates which allows to lower dislocation density in epitaxial layers by 3-4 orders of magnitude. In this paper, the gas pressure - high temperature experimental set up with precise pressure and temperature control is presented. It is shown, that the application of programmable multi-zone furnaces at hydrostatic gas pressure allows to obtain useful information concerning phase diagrams of materials (for example by Thermal or Differential Thermal Analysis) in large pressure (up to 2 GPa) and temperature (up to 2000K) range. From the analysis of thermodynamical properties of AlN, GaN and InN, which is shortly summarized in the paper follows, that the best conditions for crystal growth at available pressure and temperature conditions can be achieved for GaN. The presented experimental system can be used for the growth of high quality crystals of this material. At present both n-type and semi-insulating substrate crystals with surface area up to 1 cm2, with dislocation density of 103 - 105 cm-2 are obtained and successfully used for homoepitaxy by MOCVD and MBE. In this paper, the results of crystallization of substrate quality GaN crystals obtained with the use of large volume high pressure reactor will be presented. It will be shown that the quality of GaN crystals does not deteriorate with the increasing size and that epi-ready surfaces of GaN substrates can be obtained.

2:00 PM *DD7.2 
HIGH PRESSURE STUDY OF III-NITRIDES AND RELATED HETEROSTRUCTURES. W. Shan, J.J. Song, Oklahoma State Univ, Center for Laser and Photonics Research, Stillwater, OK; Z.C. Feng, M. Schurman, R.A. Stall, EMCORE Corporation, Somerest, NJ.

We present the results of spectroscopic study of GaN, InGaN alloys, and related heterosturctures under hydrostatic pressure using the diamond-anvil-cell technique. The predominant near-band-edge exciton emission lines of GaN were found to shift almost linearly toward higher energy with increasing pressure. While the broad emission band at the yellow spectral region showed a blue shift behavior under applied pressure, a relatively strong sublinear pressure dependence was observed. The emissions associated with the band edges of InGaN alloy samples and GaN/InGaN quantum well structures exhibited a pressure dependence very similar to the exciton emissions in GaN. We found that the linear pressure coefficient for InGaN emissions depends on the alloy composition, and shows a small decrease as the In content increases. In addition, the strain effects on the excitonic transitions in GaN have been examined. The values of the principal deformation potentials of the direct band gap for GaN have also been determined.

2:30 PM *DD7.3 
OPTICAL STUDIES OF SEMICONDUCTORS AT LARGE HYDROSTATIC PRESSURES. E.E. Haller, and M.D. McCluskey*, Lawrence Berkeley National Laboratory and University of California, Berkeley, CA.

Among the various external disturbances used in the study of semiconductors, including electric and magnetic fields as well as uniaxial stress, large hydrostatic stresses can be employed to induce dramatic changes in host lattice, dopant and defect properties. Diamond anvil cells with an appropriate pressure medium (e.g. liquid N2 or alcohol mixtures) allow the application of stresses up to hundreds of kbar. In this pressure range the global conduction band minimum can become a local minimum. GaAs for example changes near 45 kbar from a direct (-band) to an indirect (X-band) semiconductor. Donors in GaAs and InP transform from their shallow, hydrogenic state to the DX configuration at hydrostatic pressures near 23 and 82 kbar, respectively. This donor configuration change has been studied using local vibrational mode (LVM) spectroscopy in the far infrared region of the electromagnetic spectrum. Recently we have investigated several LVMs of H-containing complexes in GaAs as a function of hydrostatic pressure at liquid He temperatures. Depending on the specific complex we find the LVM frequencies to vary either linearly, sub-Ýor superlinearly on hydrostatic pressure. In the case of O in Si the vibrational mode changes its character from that of a harmonic oscillator to a rotor as pressure is applied. We will discuss the implications of the pressure dependences of LVMs.

3:00 PM *DD7.4 
A STUDY ON BAND ALIGNMENT IN GaAs/GaInP (PARTIALLY ORDERED) HETEROSTRUCTURE WITH HIGH PRESSURE. K. Uchida, P. Y. Yu*, J. Zeman+, Z. P. Su*, S. H. Kwok*, G. Martinez+, K. L. Teo*, T. Arai** and K. Matsumoto**, Dept of Communications and Systems, The Univ of Electro-communications, Chofu, Tokyo, JAPAN; *Dept of Physics, Univ of California at Berkeley and Materials Science Div, Lawrence Berkeley Lab, Berkeley, CA; +Grenoble High Magnetic Field Lab, MPI-FKF/CNRS, Grenoble, FRANCE; **Nippon Sanso Co, Tsukuba Lab, Tsukuba, Ibaraki, JAPAN.

It is well-known that the band gap of Ga0.54In_0.5P (GaInP) grown pseudomorphically on GaAs is reduced as a result of Cu-Pt type ordering. It has been found that this ordering can be controlled to some extend via the growth parameters. Since this ordering is expected to affect also the band alignment of the heterostructure,a key factor in tailoring the heterostructure of this system is to control the ordering. In this talk, we will present an overall summary on the study of the optical properties of GaAs/GaInP(partially ordered) heterostructures using high pressure tecniques. The heterostructures used in our study were grown by metal-organic vapor phase epitaxity under different conditions to vary the degree of ordering. The optical properties are studied via photoluminescence PL from GaAs/GaP/GaInP(partially ordered) single quantum well and from deep emission bands (DB) and the upconversion of photoluminescence(UPPL) at the GaAs/GaInP(partially ordered) interface. In case of the GaAs/GaP/GaInP(partially ordered)single quantum well the band alignments at the GaAs/GaP and GaP/GaInP(partially ordered) heterojunctions have been determined. High pressure and high magnetic field techniques have also demonstrated the importance of both a type II band alignment and of defect centers localized near the interface to UPPL. Finally, the pressure dependence and time-resolved PL of the DB demnstrate conclusively that this emission is due to recombination of distant donor acceptor pairs near the GaAs/GaInP interface. Our conclusion is that the observation of both UPPL and of DB are related to the higher degree of ordering in the GaInP layers.

3:45 PM *DD7.5 
HIGH PRESSURE MEASUREMENTS AT LOW TEMPERATURES AND HIGH MAGNETIC FIELDS. S.W. Tozer, National High Magnetic Field Laboratory, Tallahassee, FL.

Pressure can be used to expand the parameter space available in almost any experiment. Unlike the use of chemical alloy series, the use of pressure permits the investigation of a material's intrinsic properties by altering the interatomic or intermolecular distances within a single specimen. Problems inherent to chemical alloy studies such as batch-to-batch variation and the introduction of impurities and associated scattering are thus avoided. Pressure allows the experimenter to continuously tune the electrical and optical properties of a material. When combined with low temperatures and high magnetic fields, pressure becomes a powerful tool for the exploration of the band structure and defect levels in semiconductors and exotic transport mechanisms in molecular conductors and high temperature superconductors. Three limitations of the high pressure pulsed field experiment are the narrow bore of the magnet which dictates the use of very small diamond-anvil/sapphire ball cells to generate the pressure; the short time to and at peak field; and eddy current heating. Methods and designs to generate hydrostatic pressure to perform optical and electrical measurements in both pulsed and DC fields will be described and recent results will be presented.

4:15 PM *DD7.6 
MAGNETIC SUSCEPTIBILITY MEASUREMENTS OF SUPERCONDUCTING TC OF NIOBIUM AT MEGABAR PRESSURES *. Viktor V. Struzhkin, Geophysical Laboratory and Center for High Pressure Research, Carnegie Institution of Washington, Washington, DC.

We have measured the superconducting Tc(P) of Nb using a highly sensitive magnetic susceptibility technique to megabar pressures (> 100 GPa), and Tc(P) of Ta up to 45 GPa. We observed anomalies in Tc(P) for Nb at 5-6 GPa and 60-70 GPa, at which pressures Tc increases by 0.7 K and decreases by about 1 K, respectively. In contrast, Tc in Ta remains nearly constant up to 45 GPa. We suggest that the anomalies in Nb arise from stress-sensitive electronic topological transitions; that is, changes in Fermi surface topology and consequently in the density of states at the Fermi level are responsible for the observed changes in Tc. Between 70 and 132 GPa, Tc for Nb drops continuously to 4.7 K, which is related to the decrease in density of states at the Fermi level with increasing pressure. The experimental results are compared with existing band structure calculations for Nb at normal and reduced lattice spacings.

4:45 PM *DD7.7 
MAGNETISM, ELECTRONIC PROPERTIES, AND STRUCTURE AT HIGH DENSITY STATE OF MAGNETIC SOLIDS. A.G. Gavriliuk, G.N. Stepanov, V.A. Sidorov, I. Trojan, Institute for High Pressure Physics, Troitsk, RUSSIA, and I.S. Lyubutin, Institute of Crystalography, Moscow, RUSSIA.

We developing and applying the high pressure technique to study the modification of electronic structure, magnetic properties and local crystal structure in magnetics under high pressure. The variation in inter-atomic distances and atomic volumes under the high pressure regime will provide new information for elucidating of electronic processes in solids and understand the connection between electronic structure, magnetic and electric properties of matter. As model materials we have used rare-earth orthoferrites which are of the perovskite-type structure and Heusler alloys, cubic structure. Those materials are know to be magnetic insulators and metals respectively. We studying the following properties induced by high pressure: electronic structure, spin cross-over effects, magnetic to non-magnetic transitions, and insulator-to-metal transitions induced by high pressure. The problem of delocalization and distribution of spin and charge density in dielectric materials, semiconductors and metals, as well as chemical bonding can be tackle in this way. The Mossbauer Sn-119 spectroscopy, optical absorption spectroscopy, magnetic, DTA and compressibility measurements at high pressure were applied as tools for the investigations. High pressure Mossbauer spectroscopy investigations have been carried out in the rare earth orthoferrites NdFeO3 and LuFeO3 at pressures up to 30 GPa and in Heusler alloys Co2MnSn and Ni2MnSn at pressures up to 12 GPa at room temperatures. Measurements of pressure dependencies of Neel and Curie temperatures, compressibility measurements have been carried out at pressures up to 9 GPa. High pressure optical absorption investigations in LuFeO3 have been carried out at pressures up to 62 GPa and reviled jump in absorption edge near the 45 GPa related to the electronic transition. In these materials for the first time explained the high pressure behaviour of the hyperfine magnetic fields at the nuclei of the diamagnetic atoms and covalency effects at high density state. The experimental results obtained will allow to improve the existing theories and predict properties of newly synthesized materials.

SESSION DD8: DENSE SOLIDS - MOLECULAR AND METALLIC 
Chair: Thomas S. Duffy 
Thursday Morning, December 4, 1997 
Staffordshire (W)

8:30 AM *DD8.1 
MOLECULAR SYSTEMS UNDER HIGH PRESSURES: THEORY AND COMPUTER SIMULATIONS. Guido L. Chiarotti, Intl School for Advanced Studies, Trieste, ITALY.

Understanding the behaviour of molecular systems under high pressure and/or temperature conditions has implications in fundamental physics (pressure induced metal-insulator or magnetic-non magnetic transitions), in technology (polymerization reactions and amorphous formation upon compression), and planetary physics. Computer simulations represent a very powerful tool for the investigation of the properties of matter under these extreme conditions. Since the stability of different molecular phases is related to subtle intramolecular interactions, the use of a first principles approach is mandatory. We have recently developed a deformable-cell method for first principles molecular dynamics which allows the simulation of structural phase transformations with the correct quantum-mechanical description of interatomic forces and stress [1]. In this talk I will discuss results obtained with this methodology on a variety of simple molecular systems which include: i) the determination of the ground state structure of the broken symmetry phase (BSP) of in the pressure range 100-150 GPa [2], ii) the behaviour of methane [3] and ammonia [4] along the isentrope of the middle ice layers of Neptune (i.e. for pressures in the range 100-300 GPa and temperatures of 4000-5000 K), iii) the phase transformation in which is possibly related to the collapse of the antiferromagnetism of the phase [5].

9:00 AM *DD8.2 
AB INITIO SIMULATION OF SOLID STATE POLYMERIZATION OF ACETYLENE UNDER PRESSURE. Marco Bernasconi, Dept of Material Science, University of Milan, Milano, ITALY.

The phenomenon of polymerization is of outmost technical and scientific importance. In the vast majority of cases the polymerization is induced by chemical means in the fluid phase. An attractive alternative to this approach is to induce the polymerization in the solid state by physical means such as temperature, light and pressure. In this contribution we will show that ab initio simulations can provide crucial insight into these processes. We will present the simulation of the solid-state polymerization reaction of acetylene under pressure,1 recently detected experimentally.2 This is achieved by using the recently developed ab initio constant pressure molecular dynamics simulation scheme.3 We have found that a triplet exciton self trapped on a single, cis-bent molecule in crystalline acetylene is a very effective polymerization seed at the experimental polymerization pressure and we therefore predict that injection of triplet excitons would greatly enhance the polymerization rate. Under further compression polyacetylene undergoes a gradual saturation of C-C bonds via chain interlinks. We have found that crystalline polyacetylene converts into a-C:H containing 80 sp3 carbon atoms at 50 GPa.4 The final a-C:H is a wide gap insulator and, at variance with the conventionally generated a-C:H, is highly anisotropic keeping some memory of the original polyacetylene chains axis.

9:30 AM DD8.3 
Withdrawn.

10:00 AM *DD8.4 
PHASE TRANSITIONS IN Fe and Co AT HIGH PRESSURES AND TEMPERATURES. C.S. Yoo, H. Cynn, Lawrence Livermore National Laboratory, Livermore, CA.

The hcp-to-fcc phase transitions of Fe and Co have been studied at high pressures and temperatures by using in-situ synchrotron x-ray diffraction coupled with a diamond-anvil cell laser-heating technology. The hcp-to-fcc transitions of Group VIll elements occur systematically at high temperatures at high pressures. For example, the T/DeltaP slope of the transition rapidly decreases from Fe to Co to Ni. Consequently, the fcc phase of -Fe disappears at the liquid/fcc/hcp triple point near 5010 GPa and 2600200 K; whereas, the fcc phase of -Co is stable at high temperatures at high pressures well above 100 GPa. The fcc phase of Ni is stable even at the ambient temperature at high pressures. The high temperature fcc phases of both -Co and -Fe are quenchable at the ambient temperature. On the other hand, the fcc-to-hcp transitions in Co and Fe are rather complicated at low pressures respectively below 60 GPa and 40 GPa due to the metastable dhep formed during the transitions. In this paper, we will present the x-ray data of iron and cobalt and, then, discuss about the systematic of the hcp-to-fcc transitions in Group VIII elements and the metastability of the dhcp phases.

10:30 AM *DD8.5 
NOVEL STRUCTURE OF MgSe IN THE MEGABAR REGIME: POSITIONAL PARAMETER DETERMINATION TO 150 GPa. Arthur L. Ruoff, Ting Li, Chandrabhas Narayana, Huan Luo and Raymond G. Greene, Cornell University, Department of Materials Science and Engineering, Ithaca, NY.

A review of the transitions at high pressure in the alkaline chalcogenides from the six-fold NaCl structure to the eight-fold CsCl structure will be given along with predictions for this transition in the magnesium and beryllium compounds. In energy dispersive x-ray diffraction studies on MgSe at CHESS we found that it transformed instead to a seven-fold cordinate structure (a modified NaCl structure) which is the Fe-Si or Au-Be structure. This second-order transformation begins at 100 GPa. At 148 GPa, were the transition appears to be complete, we have accurately determined the positional parameters u and w to within 0.002. This is the first case in which positional parmeters have been determined in a structure found above 100 GPa.

11:00 AM DD8.6 
IMAGE PLATE X-RAY DIFFRACTION STUDY OF DISTORTED FCC PHASE IN RARE EARTH METALS AT HIGH PRESSURES. Yogesh K. Vohra, Steven Beaver, Gary Chesnut, Dept of Physics, University of Alabama at Birmingham (UAB), Birmingham, AL.

It is well established that the rare earth metals undergo a sequence of phase transformations, i.e., hcp to Sm-type structure to dhcp to fcc structure with increasing pressure. The structural transitions in rare earth metals are driven by s to d electron transfer at low pressures and f-delocalization at high pressures. Also, the high pressure fcc-phase in rare earth metals is known to further transform in to a low symmetry distorted fcc-phase in a subtle and continous manner with increasing pressure. There have been many proposals for the crystal structure of this distorted fcc phase but conclusive structural assignments have not been made. One of the fundamental reasons behind this is the lack of high resolution x-ray diffraction data on rare earth metals at high pressure. In our study, we employed high resolution image plate x-ray diffraction technique using F2 wiggler line at CHESS, Cornell University using monochromatic x-ray energy of 40.443keV. The prototype rare-earth metal selected for this study was Gadolinium (Gd) where a transition from fcc to distorted fcc phase occurs at 33 GPa. In our case, image plate x-ray diffraction pattern of the distorted fcc phase was obtained at 47 GPa. We will compare fits to various structural proposals like triple hexagonal close packed(thcp), trigonal cell with 6 atoms and a hexagonal cell with 24 atoms suggested in the literature. We will also discuss the implications of the present results on the f-delocalization phenomenon in rare earth metals which is usually attributed to the appearance of the low symmetry crystal structures at ultra high pressures.

11:15 AM DD8.7 
PRESSURE INDUCED ANOMALIES IN c/a AND ELECTRIC FIELD GRADIENT AND ELECTRONIC TOPOLOGICAL TRANSITIONS IN Zn AND Cd. D.L.Novikov, A.J.Freeman, Science and Technology Center for Superconductivity, Northwestern Univ., N.E.Christensen, A.Svane, Aarhus Univ., DENMARK; C.O.Rodriguez, IFLYSIB, ARGENTINA; M.I.Katsnelson, Institute of Metal Physics, Yekaterinburg, RUSSIA.

Recent pressure experiments1 found an anomalous behaviour of the c/a ratio and the Lamb-Mössbauer factor for hcp Zn. We simulate the effect of hydrostatic pressure on the electronic structure, lattice parameters and electric field gradients (EFG) for hcp Zn and Cd using the full-potential linear muffin-tin orbital method2 in conjunction with the new Perdew-Burke-Ernzerhof generalized gradient approximation (PBA-GGA)3 to the density functional for exchange-correlation. Theoretical equilibrium volumes for Zn and Cd are found to be in excellent agreement with experiment (whereas non-GGA corrected LDA underestimates them by as much as 10%). We do find the anomaly in the pressure dependence of c/a at reduced unit cell volumes (V/V0=0.88 for Zn and 0.85 for Cd) and a similar anomaly in the EFG tensor. The analysis of the Fermi surface transformations in these metals shows a number of electronic topological transitions (ETT) occuring with pressure. One of them, the appearance and fast growth of needles along the K-H direction in the Brillouin zone, coincides with the appearance of the anomaly in c/a and might be a reason for it. At the same time we do not confirm that a collapse of the giant Kohn anomaly is related to the c/a anomaly as suggested earlier. Finally, we investigated the influence of the pressure induced ETT's on the low-temperature thermoexpansion coefficiens.

11:30 AM DD8.8 
PRESSURE-INDUCED PHASE TRANSITION IN FeI2 AND IMPLICATIONS FOR THE INSULATOR-METAL TRANSITION. M.B. Kruger(1), J.H. Nguyen(2), Y.M. Li(1), W.A. Caldwell(3), L.R. Benedetti(4), A. Kavner(3), and R. Jeanloz(3). (1) Department of Physics, University of Missouri, Kansas City, MO; (2) Lawrence Livermore National Laboratory; (3) Department of Geology and Geophysics, University of California, Berkeley, CA; (4) Department of Physics, University of California, Berkeley, CA.

We have studied the structure of the transition-metal compound FeI2 under high pressures using angle dispersive x-ray diffraction from 0-62 GPa, well above the previously reported insulator-metal transition pressure of 23 GPa. We observed a discontinuous change in volume of approximately 10%, at 23 (±2) GPa indicating a first order phase transition, which is correlated with substantial changes in the isomer shift and quadrupole splitting observed in a previous M–ssbauer study. The volume discontinuity suggests that the qualitative characteristics of the ambient pressure wave functions and energy bands are not suitable for interpreting the high pressure insulator-metal transition for FeI2, as has been done previously for NiI2. We propose instead a possible new mechanism for metallization due to conduction on two-dimensional iodine planes. Like related transition metal iodides, FeI2 is relatively soft with a zero pressure bulk modulus and pressure derivative of K0 = 23 (±4) GPa and K0' = 3.8 (±0.5), respectively.

11:45 AM DD8.9 
A HIGH PRESSURE RAMAN STUDY OF THE FERMI RESONANCE AND PHASE TRANSITIONS IN CuCl. C. Ulrich, A. Gobel, K. Syassen, and M. Cardona, Max-Planck-Institut fur Festkorperforschung, Stuttgart, GERMANY.

We have measured Raman spectra of CuCl (isotopically pure samples 63CuCl and 65CuCl) under hydrostatic pressure up to at low temperature (). For the ambient pressure zinc-blende structure the Raman line shape of the transverse-optic (TO) mode of CuCl is anomalous, consisting of a broad structure with two maxima. This feature can be explained by the strong anharmonic interaction of the TO mode with two-phonon combination bands of acoustic modes (Fermi resonance). Pressure is a means to tune the phonon frequencies and thus the anharmonic interactions. This results in a drastic change of the measured Raman spectrum near the TO mode. Based on a shell model for the lattice dynamics of CuCl we have successfully simulated the effect of pressure on the anomalous TO phonon line shape and the effect of isotopic substitution on the pressure dependence of the Raman spectra. We have also examined the Raman spectra of semiconducting high pressure phases of CuCl. The latter results are used to map out the phase stability regimes of the high pressure phases at low temperature.

SESSION DD9: METASTABILITY, AMORPHIZATION AND GLASSES 
Chair: Stefano Baroni 
Thursday Afternoon, December 4, 1997 
Staffordshire (W)

1:30 PM *DD9.1 
METASTABLE MELTING LINES OF ICE PHASES AT LOW TEMPERATURES. O. Mishima, Natl. Inst. Res. Inorgan. Mat., Tsukuba, JAPAN; H. E. Stanley, Boston Univ, Center for Polymer Studies and Dept of Physics, Boston, MA.

When ice Ih in an emulsion is compressed below 250K, it melts to supercooled liquid water, avoiding the formation of other crystal phases[1]. Recently, we created emulsified high-pressure ices under high pressure and low temperature in a piston-cylinder apparatus, and measured their temperature while these ices were decompressed at a constant rate at different temperatures. We could detect metastable melting points of high-pressure ices, and could identify their melting lines to low temperatures in the pressure-temperature phase diagram of ice. Here, we discuss the following results. 1) the location of the homogeneous nucleation temperature line separating supercooled liquid water and high-pressure ices up to about 15 kb, 2) the identification of what could be one and possibly two new ice phases, 3) the relationship between decompression-induced melting and decompression-induced amorphization, 4) evidence at high pressures that is not consistent with the Speedy-Angell hypothesis, 5) abnormal behavior of the melting line of ice IV around 215 K and 1 kb that is consistent with the hypothesis of a liquid-liquid phase transition.

2:00 PM *DD9.2 
PRESSURE-TUNING CRYSTAL MORPHOLOGY AND DIELECTRIC DOMAINS. P.F. McMillan, H. Hubert, L.A.J. Garvie, B. Devouard, P. Buseck, A. Grzechnik*, A. Chizmeshya, G.H. Wolf, W.T. Petuskey, MRSEC, Dept of Chemistry, Arizona State University, Tempe, AZ. *Now at ENS-Lyon, FRANCE.

The effect of pressure to force ions such as Ge4+ and Si4+ into octahedral coordination in perovskites is well known. Decompression of these materials results in lattice instabilities which can lead to amorphization, or recovery in a highly metastable state. The instability leading to amorphization results from an underlying phonon instability which resembles the ferroelectric distortion encountered in titanate perovskites. We have used a combined theoretical and experimental approach to prepare silicate and germanate perovskites and solid solutions (e.g. SrGeO3-SrTiO3) which exhibit high values of their dielectric constant when decompressed to room pressure. The combination of ferroic and elastic instabilities results in a dispersion of domain sizes, which may lead to relaxor behavior in a chemically homogeneous system. In work on the B-O system, there had been previous reports of phases with the stoichiometry B2O prepared at high pressure with both the graphite and ordered diamond structures. In synthesis experiments designed to repeat this work, we found that the stable phase was in fact the well known (but as yet poorly characterized) material B6-xO [10]. This material is of interest because of its high hardness, and its itrinsic red color indicates potentially interesting semiconducting properties. The high pressure syntheses are conducted in parallel with electron energy loss techniques which permit precise compositional analysis of the light-element bearing compounds in conjunction with microstructural characterization. In syntheses at 5-6 GPa, a remarkable result was found: the B6O formed 10-20 mm particles with nearly perfect icosahedral morphology. These have potential applications in high wear environments.

2:30 PM *DD9.3 
PRESSURE AS A PROBE OF THE GLASSY STATE OF FERROELECTRICS WITH RANDOM SITE DISORDER. George A. Samara, Sandia National Laboratories, Albuquerque, NM.

There has been much recent interest in the dynamic and static properties of ferroelectrics in which randomly competing interactions lead to the formation of a glassy (or Relaxor) state at low temperature. The usual manner of studying these properties has been to very the composition and degree of disorder to induce the glassy state. However, this approach introduces added randomness, compositional fluctuations, lattice defects, and changed interatomic forces which complicate the interpretation of results. Pressure turns out to be a much cleaner variable for studying such systems. By applying pressure to a sample of fixed compositions, one varies only the interatomic interactions and balance between long- and short-range forces, making it possible to get to the essential physics. This presentation will illustrate the power of this approach. We have investigated mixed ABO3 perovskites of the PbZr(1-x)TixO3 family with additional substitutions at both the A and B sites. Results have revealed: a pressure-induced crossover from normal ferroelectric- (nFE-) to-relaxor (R) behavior; the continuous evolution of the dynamics and energetics of the relaxation process; a spontaneous R-to-nFE transition at a temperature well below the dynamic glass temperature of the R phase; and the vanishing of this transition with pressure at a critical point. These results can be understood in terms of a large decrease in the correlation radius among polar nanodomains - a unique property of soft phonon mode ferroelectrics.

3:00 PM DD9.4 
PRESSURE-INDUCED AMORPHIZATION AND DISORDERING ON COOLING IN SEMI-CRYSTALLINE POLYMERS. S. Rastogi, G. Hoehne, A. Keller, P.J. Lemstra, Eindhoven Polymer Laboratories, Eindhoven University of Technology, Eindhoven, NETHERLANDS; Department of Physics, University of Ulm, Ulm, GERMANY; H.H. Wills Physics Lab., University of Bristol, Bristol, UNITED KINGDOM.

In the course of exploring the phase diagram of the polymer poly(4-methyl-pentene-1) and syndiotactic polystyrene as a function of temperature and pressure by in-situ X-ray diffraction, in-situ Raman Spectroscopy and high pressure DSC, we have discovered some unusual phase behaviour. The polymer, crystalline under ambient conditions, becomes amorphous on increasing pressure in two widely separated regimes,i.e. below the glass transition temperature (20C) and below the melting temperature (200C). This suggests the possibility of re-entrant of two widely separated amorphous phases as pressure or temperature are varied. In the higher temperature regime, the melting point shows a maximum as a function of pressure. The lower-temperature amorphous phase becomes crystalline on heating, and reverts to the glassy disordered phase on cooling. Inversion in the melting temperature, amorphization below the glass transition temperature and disordering on cooling have been confirmed with the help of high pressure DSC. Similar transitions have been found by us in another polymer, syndiotactic polystyrene. The pressure-induced amorphization has been observed previously in other systems, such as ice, silica and phosphates.

3:45 PM DD9.5 
HIGH PRESSURE CARS AND PHOTON ECHO STUDIES OF MIXED CRYSTALLINE AND AMORPHOUS MOLECULAR SOLIDS. Eric L. Chronister, Bruce J. Baer, Otto Berg, University of California, Riverside, Dept of Chemistry, Riverside, CA.

High pressure studies of vibrational relaxation and electronic dephasing in mixed molecular crystals and glasses have been performed using picosecond CARS and photon echo measurements in a diamond anvil cell. Picosecond CARS measurements have been used to probe the effect of pressure on vibrational relaxation in neat molecular crystals, and photon echo measurements have been utilized to determine the effect of high pressure on the homogeneous electronic dynamics of chromophores in inhomogeneously broadened crystalline and amorphous solids. High pressure photon echo results are presented for mixed crystals (e.g. pentacene doped p-terphenyl and naphthalene), as well as doped organic polymers (e.g. rhodamine and pentacene doped PMMA and polystyrene). High pressure photon echo results in mixed crystals reveal pressure induced homogeneous line narrowing. The line narrowing observed at high pressure occurs due to either a pressure induced increase in the guest pseudolocal phonon frequency (resulting in a decrease in the thermal occupation), or due to an increase in the pseudolocal phonon lifetime. Temperature dependent (1-10 K) photon echo measurements yield both the lifetime and the frequency associated with the guest pseudolocal phonon. The high pressure photon echo measurements probe the anharmonicity of local phonon dynamics and they provide a unique probe of the dynamics associated with pressure induced phase changes of the host lattice. The effect of pressure on the low frequency dynamics in doped glasses has also been investigated. Temperature dependent photon echo measurements under variable high pressure conditions have been used to investigate the low frequency dynamics of amorphous polymers, typically characterized by tunneling two level system (TLS) models. The effect of pressure on the low frequency dynamics in doped amorphous polymer systems are contrasted with the local phonon behavior observed in mixed crystalline systems.

4:00 PM *DD9.6 
PHASE TRANSFORMATIONS OF BERLINITE-TYPE FePO4 UNDER PRESSURE, M.P. Pasternak1, G. Kh. Rozenberg1, E. Milner1, M. Amanowicz1, T. Zhou2, U. Schwarz2, K. Syassen2, R. Dean Taylor3, M. Hanfland4, and K. Brister5. 1 School of Physics and Astronomy, Tel Aviv University, Tel Aviv, Israel 
2 MPI für Festkörperforschung,Stuttgart, Germany 
3 Los Alamos National Laboratory, Los Alamos NM 87545, USA 
4 ESRF, Grenoble, France 
5 CHESS, Wilson Lab., Cornell University, Ithaca NY.

We have investigated the high pressure behavior of berlinite-type FePO4 by x-ray diffraction, Raman scattering, and Mössbauer spectroscopy. A simultaneous onset of disordered and high pressure crystalline phases with about equal abundance is observed at a pressure of 2.5(5) GPa. The structure of the new crystalline phase is identified as VCrO4-type. In this phase the Fe ions are sixfold coordinated, which is also the predominant Fe coordination for the amorphous component. A small amount of Fe in fourfold coordination is retained up to at least 25 GPa. The coexisting crystalline and amorphous components show stable relative abundance to these pressures, and persist upon fully releasing the pressure at 300 K. These phenomena of concurrent amorphous and crystalline transformation at low hydrostatic pressure and stable abundance ratio over a large pressure-range are unique in pressure-induced structural transformations of SiO2 analogues.

4:15 PM DD9.7 
VANISHING ATOMIC MIGRATION BARRIER IN SILICA. Michael J. Aziz, Division of Engineering and Applied Sciences, Harvard University, Cambridge, MA; Susan Circone, Department of Earth and Planetary Sciences, Harvard University, Cambridge MA; Carl B. Agee, Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA.

Pressure (1-3 GPa) enhances the rate constants for atomic transport such as diffusion, viscous flow, and crystal growth in silica and some silicate melts. Structural transitions and coordination changes observed beyond 10 GPa may be related to this pressure-induced increase in atomic mobility. However, there must be limits on how much pressure can enhance the mobility. A migration barrier decreasing linearly with pressure should vanish, creating a free energy catastrophe at a critical pressure Pc, beyond which a sudden change in behavior should be observed. Here we report measurements of the pressure-dependence of the growth rate of quartz from amorphous silica at 1673 K. The sharp peak in growth rate at 3 GPa is evidence that Pc is being traversed. An atomistic mechanism to interpret this behavior is presented.

4:30 PM DD9.8 
PRESSURE INDUCED AMORPHIZATION IN BAs: A POSSIBLE INHIBITED DISSOCIATION. Renata M. Wentzcovitch, Dept of Chemical Engineering, University of Minnesota, Minneapolis, MN.

Amorphization in the III-V semiconductor BAs occurs at approximately 125 GPa. This pressure is close to the predicted transition pressure for the zincblende to rocksalt transition ( 110-120 GPa at at T = 0 K). Pressure induced amorphization in general occurs in phases kept metastable far beyond their thermodynamical stability field, when the transformation to the stable phase is kinetically inhibited at low temperatures. The near coincidence of the amorphization pressure with the predicted thermodynamical transition pressure seems to be a unique case. We have investigated the possibility of a pressure induced dissociation at T = 0 K, similar to the peritectoide reaction (B6As + As <=> BAs) which occurs at 1070 C, and suggest that this inhibited phenomenon is a possible underlying cause of the ``amorphization.''