Abstracts Physics

High temperature phase equilibria studies in the Bi-Sr-Ca-Cu-O-Ag system

by Lawrence Margulies

A variety of experimental techniques were utilized to examine the high temperature phase equilibria in the Bi-Sr-Ca-Cu-0-Ag system. Quenching studies were used to determine the liquid solubility of Ag in the Bi2Sr2CaCu2O8 (Bi2212) melt and the details of the peritectic decomposition pathway of Bi2212 as a function on Ag content and oxygen partial pressure (PO2). A liquid immiscibility region between oxide and Ag liquids in the 8-98 at% range was found above 9000C. Two eutectics were found in the Bi2212-Ag pseudo-binary. On the oxide rich side, a eutectic exists at approximately 4 at% Ag. On the Ag rich side, a eutectic exists at approximately 98 at% Ag at a temperature of 150C below the melting point of pure Ag. Six distinct solid phases were found to be in equilibrium with the partial melt within the Ag content and P02 range studied. The stability of these solid phases were found to be highly sensitive to P02, and to a much lesser extent Ag content. High temperature x-ray diffraction (HTXRD) studies of this system are in conflict with these results.;It is suggested that these discrepancies are due to experimental artifacts caused by the significant thermal gradients and lack of full bulk sampling which is inherent in conventional HTXRD designs. In part II, a new furnace design compatible with synchrotron radiation sources is introduced to address these problems. This design allows for full bulk sampling in a low thermal gradient environment using Debye-Scherrer transmission geometry. Sample spinning is also introduced in the design to eliminate preferred orientation and incomplete powder averaging and allow for quantitative phase analysis and structural refinement. Studies on model systems are presented to demonstrate the capabilities for high resolution structural studies (Al2O3) and time resolved phase transformation studies (SrCO3). Finally, the Bi2212 system is examined to confirm the quenching results of part I, and to demonstrate the degree to which this new HTXRD design solves the problems associated with conventional designs.