The structure of liquid Sn was studied by neutron scattering experiments in the widest temperature range that was ever performed. Though, on increasing temperature, the existence of the shoulder in the structure factor, S(Q), becomes less clear in the change of the overall shape of the S(Q), the structure related to this shoulder seems to be present even at 1873 K. The first-principle molecular-dynamics ͑FPMD͒ simulation was performed for the first time for liquid Sn by using the cell size of 64 particles. The calculated results well reproduced S(Q) obtained by the neutron experiments. The angle distribution, g (3) (,r c ), was evaluated for the angle between vectors from centered atom to other two atoms in spheres of cutoff radii r c 's. The g (3) (,r c ) shows that, with the decrease of r c from 0.4 to 0.3 nm, a rather sharp peak around 60°disappears and only a broad peak around 100°remains; the former peak may be derived from the feature of the closely packed structures and the latter one is close to the tetrahedral angle of 109°. In addition, the coordination number, n, of liquid Sn counted within the sphere of r c ϭ0.3 nm is found to be 2-3 and does not change with the increase of temperature even up to 1873 K. These facts indicate that at least the fragment of the tetrahedral unit may be essentially kept even at 1873 K for liquid Sn. For comparison, the FPMD simulation was performed for the first time also for liquid Pb. No sign of the existence of the tetrahedral structure was observed for liquid Pb. Unfortunately, the self-diffusion coefficients, D's, obtained from this FPMD for liquid Sn do not agree with those obtained by the microgravity experiments though the structure factors, S(Q)'s, are well reproduced. To remove the limitation of the small cell size of the FPMD, the classical molecular-dynamics simulations with a cell size of 2197 particles were performed by incorporating the present experimental structural information of liquid Sn. Obtained D's are in good agreement with the microgravity data.

A 2-mm-diameter glass sphere of ferroelectric BaTi2O5 was fabricated from melt using containerless processing. The glass structure was analyzed by high-energy X-ray diffraction using an incident photon energy of 113.5 keV, indicating that distorted Ti−O polyhedra, with average coordination number (N Ti - O) of approximately 5, presented in the glass. Above the glass transition temperature (972 K), three successive phase transitions, from glass to a metastable α phase at 972 K, then to a metastable β phase at 1038 K, and finally to a stable monoclinic γ phase above 1100 K, were observed. At the crystallization temperature of the α phase, the permittivity jumped instantaneously by more than 1 order of magnitude, reaching a peak of 1.4 × 107. This interesting phenomenon, occurring near the crystallization temperature, has important technical implications for obtaining an excellent dielectric glass−ceramic through controlled crystallization of BaTi2O5 glass.

Metallic liquid silicon at 1787K is investigated using x-ray Compton scattering. An excellent agreement is found between the measurements and the corresponding Car-Parrinello molecular dynamics simulations. Our results show persistence of covalent bonding in liquid silicon and provide support for the occurrence of theoretically predicted liquid-liquid phase transition in supercooled liquid states. The population of covalent bond pairs in liquid silicon is estimated to be 17% via a maximally-localized Wannier function analysis. Compton scattering is shown to be a sensitive probe of bonding effects in the liquid state. Silicon (Si) presents a fascinating phase diagram as is the case in other systems that form tetrahedrally coordinated networks.[1] Upon melting, Si transforms into a metal accompanied by a density increase of about 10%. The resistivity of liquid Si (l-Si) at the melting temperature T m is 0.75 µΩm, which is comparable to that of simple liquid metals such as l-Al. However, the first neighbor atomic coordination number in l-Si remains 5.5∼6 [2], which is approximately half that of simple liquid metals, hinting that covalent bonds survive even in the metallic state [3]. In fact, molecular dynamics simulations of molten Si at 1800K suggest that approximately 30% of the bonds are covalent and that these covalent bonds possess a highly dynamic nature, forming and breaking up rapidly on a time scale of 20 fs [4]. It is remarkable that two completely different types of bonds−metallic and covalent− can coexist in l-Si. In fact, the coexistence of two forms of liquid in a single component substance has been predicted to undergo a phase transition as a function of temperature and / or pressure [5], and many theoretical studies support the existence of a liquidliquid phase transition (LLPT) [6][7][8]. A recent study reports that l-Si could undergo an LLPT below about 1232K and above about -12kB, separating into a highdensity metallic liquid (HDL) and a low-density semimetallic liquid (LDL) [8]. But, 1232K is far below the melting temperature of 1683K of Si, and as a result the supercooled state has remained inaccessible to current experimental techniques. Very recently, Beye et al. have performed time-resolved x-ray measurements on Si using a femtosecond pulse-laser [9] to reveal liquid polymorphs of Si which could support an LLPT, but these experimental conditions are far from being ideal [6][7][8] so that the experimental confirmation of an LLPT in Si remains an open question.A key requirement for the possibility of an LLPT obviously is that the metallic and covalent bonds coexist in l-Si. Although experimental investigations of the atomic configuration hint at the existence of covalent bonds in l-Si, surprisingly, soft x-ray [10] and magnetic susceptibility measurements [11] of electronic properties so far do not support this viewpoint in that all four valence electrons in l-Si appear to behave like free-electrons. Emissivity and thermal conductivity of l-Si also exhibit a freeelectron like temperature depen...

Glass spheres of 2 mm diameter of the title compound are prepared from mixtures of BaTiO3 and TiO2 powders by containerless processing in an aerodynamic levitation furnace (200 MPa, 1427 K, 10 h). The samples are characterized by high-energy XRD. There are three successive phase transitions above the glass transition temperature of 972 K. At 994 K, the permittivity jumps instantaneously by more than one order of magnitude, reaching a peak of 1.4·10 7 . This phenomenon has important implications for obtaining an excellent dielectric glass-ceramic through controlled crystallization of BaTi2O5 glass. It is believed that bulk BaTi2O5 glass will become an important source for fundamental physics study and for practical applications. -(YU*, J.; ARAI, Y.; MASAKI, T.; ISHIKAWA, T.; YODA, S.; KOHARA, S.; TANIGUCHI, H.; ITOH, M.; KUROIWA, Y.; Chem. Mater. 18 (2006) 8, 2169-2173; ISS Sci. Proj. Off., Japan Aerosp. Explor. Agency, Tsukuba, Ibaraki 305, Japan; Eng.) -W. Pewestorf 28-011

The microgravity condition is one of the ideal conditions for the measurements of diffusion coefficients in high temperature melts because of the absence of convection in the liquid sample. Many kinds of experimental techniques, such as the long capillary method and the shear cell method, have been devised in ground based researches for the measurement of diffusion coefficient in such high temperature melts. Recently, the shear cell technique coupled with the microgravity condition was applied to the measurements of high temperature metallic melts. This technique enables us to measure the diffusion coefficient with high precision. In this paper, a brief discussion is given about the trends of previous diffusion experiments in space performed particularly by Japanese researchers. Recently, the Japanese space agency, JAXA, has developed the shear cell technique for future space experiments in JEM. The current status of the development of our shear cell technique is summarized and new results are presented as typical experiments on the ground by using test samples, for example, liquid silver–gold alloys.

No abstract

An electrostatic levitation (ESL) furnace was designed for the neutron diffraction study of condensed matter. This apparatus is composed of an electrostatic levitator, a neutron inlet path with Cd slit, a 160 degree window for the scattered neutrons, a CO2 laser for sample heating, and an optical pyrometer for temperature measurements. A preliminary neutron diffraction experiment with this ESL was performed for polycrystalline alumina. The sample position could be controlled with an accuracy of ±0.1 mm. The observed Bragg peaks were in complete agreement with those derived from the lattice data of alumina obtained from the literature value. This indicates that this facility is attractive for the structural study of condensed matter without containers.

No abstract