Project Description:
Extending Empirical Models of Perovskites to Include Chemical Ordering
Faculty Member: Prof. Rick Ubic
Perovskites are ubiquitous in modern devices because their wide compositional range and structural variability gives rise to useful properties like piezoelectricity, pyroelectricity, ferroelectricity, superconductivity, colossal magnetoresistance, proton conduction, catalysis, and spin-dependent transport. They are also of interest for use as substrates or buffer layers for compound semiconductor heteroepitaxy and as films in tunable ferroelectric superlattice structures. While a few theoretical models exist for some perovskites, none are particularly suited to the impure, doped, or otherwise defective ceramics which abound in commercial devices. For this project, we will take an entirely novel and transformative approach to discover new crystallochemical trends in perovskite materials. The overall strategy is to generalize from observations of prototype systems and use mathematical methods to help link observable macroscopic trends to phenomena on the atomic scale. These models would allow the prediction of structures with little or no experimental data, thus eliminating much of the trial and error pitfalls experienced through empirical approaches. Doing so will drastically reduce material development time and costs — a cornerstone of the Materials Genome Initiative. The specific goal of this project is to establish a generic numerical model for the effective ionic radii and lattice constants for complex perovskites containing 1:2 B-site ordering. It has already been established that rock salt (1:1) ordering on the B site causes an overall volume shrinkage with respect to the disordered analogue. Similar behavior is expected for 1:2 ordering, but it is not yet quantifiably predictable. From the data obtained experimentally, the effective shrinking of the B cations can be calculated and a model derived.
Role of Participant(s):
Participant(s) will synthesize various compositions in the system (Ba1-xSrx)[(Mg,Zn)1/3(Nb,Ta)2/3]O3 via solid-state mixed-oxide methods and characterize them for phase-purity via x-ray diffraction (XRD). Lattice constants will be derived from Le Bail refinements of high-resolution XRD data. This model system will allow the effect of B-site ordering on effective ionic sizes and cell volume to be studied as a function of rA and therefore tolerance factor in trigonal perovskites. A data-mining approach will also be employed to include other reported perovskites with this ordering in the model. Data will be categorized and analyzed for quantifiable trends using simple mathematical tools (e.g., Mathematica).