Optimization of the sintering of Aurivillius phases using D-optimal statistical design

  • PDF / 676,042 Bytes
  • 5 Pages / 612 x 792 pts (letter) Page_size
  • 21 Downloads / 129 Views

DOWNLOAD

REPORT


FF4.8.1

Optimization of the sintering of Aurivillius phases using D-optimal statistical design M.S. Peterson and S.T. Misture New York State College of Ceramics at Alfred University, Alfred NY 14802 ABSTRACT A statistical model has been employed to determine the optimal sintering times and temperatures for Bi2Sr2Nb2TiO12. The phase of interest was made in pure form by solid-state synthesis. Bi2O3 was used as a sintering aid to decrease sintering temperatures and increase densification. The amount of Bi2O3 added and the sintering temperature had a statistical impact on density, while the heat treatment time did not. INTRODUCTION In 1949, a series of papers [1-3] were published by Bengt Aurivillius exploring the discovery of mixed-metal oxides having bismuth oxide layers alternating with perovskite structure layers. These materials have the general composition Bi2An-1BnO3n+3 (where A=Ca, Sr, Ba, Pb, Bi, Na, K, etc., and B=Ti, Nb, Ta, Mo, W, Fe, etc.). The A cations can be mono-, di-, or trivalent ions; or a mixture. The Bi2O22+ sheets are separated by perovskite-type blocks (An-1BnO3n+1)2- of variable thickness according to the value of n. Because of their ionic structural framework, Aurivillius phases exhibit great flexibility with respect to metal cation substitution; therefore, these phases have high potential for systematic control of their properties [4]. This research explored the sintering behavior of Aurivillius phases in an effort to reduce sintering times and temperatures, while increasing the density. This was accomplished by using a statistical design experiment that compared the sintering temperature, sintering time, a sintering accelerator, and the final density. The Bi2Sr2Nb2TiO12 phase was chosen because of its ability to be formed in pure form by solid-state synthesis. One of the inherent problems with the Aurivillius phases is the repeatability of producing single-phase materials. Work done by Peterson and others [5-7] have shown that many of the Aurivillius compositions, in particular the oxygen deficient compositions, are not phase pure. Bi2O3 was chosen as a sintering accelerator because of its successful use in other electroceramics [8]. PROCEDURE The first step was to make Bi2Sr2Nb2TiO12 via solid-state synthesis and check for phase purity. Stoichiometric mixtures of precursor powders (Bi2O3, Alfa Aesar, 99.975% pure; SrCO3, Alfa Aesar, 97.5% pure; Nb2O5, Alfa Aesar, 99.9% pure; TiO2, Alfa Aesar, 99.9% pure) were intimately mixed in a mortar and pestle. The powders were then calcined at 815° C for 10 hours in air. The powders were reground, pressed into pellets (22 mm in diameter) at 102 MPa, and subsequently put in a powder bed (of the same composition) in a MgO crucible. The pellets were fired twice at 1095° C for twelve hours with regrinding in between. Finally, the pellets were fired at 1095° C for 24 hours. At each regrinding step, the powders were checked for phase purity by XRD. Experimental Design Optimizer [9] was then used to design a full quadratic model that would find the effects o