Crystal stability and the theory of ferroelectricity

W. G. Cochran

Advances In Physics · 1960 · 1.5K citations · 17 references

Concepts

TL;DR

Ferroelectricity in pseudo‑cubic crystals is analyzed through normal mode vibrations. The theory extends to other cubic crystals such as barium titanate, introducing ionic polarizability and a polarizability catastrophe within lattice dynamics, and discusses additional predictions. The study shows that lattice vibration parameters can be chosen to produce ferroelectric or anti‑ferroelectric behavior, that Devonshire’s phenomenological parameters can be expressed via atomic parameters, that physically reasonable values reproduce barium titanate’s dielectric properties and atomic displacements at the cubic‑tetragonal transition, and that the lowest dielectric dispersion frequency is about 3 × 10¹¹ c.p.s., varying with temperature in the cubic phase.

Abstract

Abstract Abstract The phenomenon of ferroelectricity in pseudo-cubic crystals is discussed in terms of the normal modes of vibration. It is shown that the parameters which determine the lattice vibrations of a diatomic ionic crystal may be chosen in such a way that the crystal will exhibit ferroelectric properties, and that ferroelectric or anti-ferroelectric transitions may be regarded as the result of an instability of the crystal for a certain normal mode of vibration. The theory is extended to apply to other cubic crystals, including barium titanate, and the concepts of ‘ionic polarizability’ and ‘a polarizability catastrophe’ are discussed in terms of lattice dynamics. Certain of the parameters which appear in Devonshire's phenomenological theory of ferroelectricity are found to be expressible in terms of atomic parameters. Values for the latter which are physically reasonable are found to account quite well for the dielectric properties of barium titanate and for the relative movements of the atoms which occur at the cubic-tetragonal transition. The lowest dielectric dispersion frequency is calculated to be about 3 × 1011 c.p.s. for barium titanate, and to be a function of temperature in the cubic phase. Other predictions of the theory are discussed.

References

17