SUZUKI PRECIPITATES IN NaCl
Michael Chall, Wulf Depmeier, Björn Winkler
Mineralogisches Institut, Universität Kiel, D-24098 Kiel,
Germany
mc@min.uni-kiel.de
Keywords NaCl, precipitates, size analysis, computer modelling
When NaCl or other alkali halides are doped with salts of divalent cations, one regularly observes the formation of precipitates, often consisting of the so-called Suzuki-phases with the composition A+6B2+X8. These have an NaCl-superstructure with ordered cation-vacancies and crystallise in space group Fm3_m with twice the lattice parameter of the corresponding AB compound [1]. In most cases, it has not been possible to obtain them as pure phases. Instead, they may only exist metastably embedded in a NaCl-type matrix [2]. However, their linear dimension can be as large as several thousand A [3,4].
Here we present first results of our theoretical and experimental investigations of the Suzuki precipitates in NaCl doped with CdCl2, which we study with ab initio and force-field calculations as well as x-ray diffraction techniques. The aim of our studies is to learn about their micro-, meso-, and macroscopic structure, thus allowing us to better understand the interdependencies between structural and physical properties on these different length scales.
The crystal structure was studied with ab initio quantum mechanical calculations, using the full-potential linearized augmented plane wave code WIEN97 [5]. After testing the accuracy and precision of this approach to ionic crystals by studying the pressure induced phase transition in NaCl, we derived the previously only inaccurately known crystal structure of Na6CdCl8 and its variation with pressure. The latter is difficult, if not impossible, with conventional methods, due to the lack of single-phased samples and the superposition of half of the Suzuki peaks with NaCl peaks in multi-phased samples.
The ab initio calculations are computationally very demanding, and hence it is necessary to employ simpler calculation schemes to investigate the interactions of finite crystallites with the surrounding matrix. Based on the ab initio data, we therefore intend to parameterize the energy surface in order to perform force-field calculations. With these, it becomes possible to study realistic Suzuki-NaCl assemblies, including interfaces, on an atomic scale. First results will be presented at the forthcoming meeting.
To complement the theoretical investigations, the size distribution of the precipitates within the surrounding matrix is currently investigated with x-ray line broadening analysis. The outcome of these measurements will also be presented at the meeting.
We would like to thank the Deutsche Forschungsgemeinschaft for financial support (De412/18-1) and Peter Blaha and Karlheinz Schwarz for continuous intrest and support.