A unified mechanism for ferroelectric aging

Time-dependent changing of ferroelectric, dielectric and piezoelectric properties (so-called aging phenomena) is well observed in ferroelectrics and is an important issue both in physics and in applications. The origin of aging had been phenomenologically ascribed to domain stabilization by defects. However, a clear microscopic explanation has remained controversial for many decades.

The main controversy comes from whether the stabilization is due to domain wall pinning effect or volume effect. Recently we made a single-domain Mn-doped BaTiO₃ single crystal and studied its aging effect. By this critical experiment, we solved the long-standing puzzle, providing a unified explanation for all ferroelectric aging.

Experimental results showed that: aged single domain sample has a peculiar double P-E hysteresis loop, contrasting with the normal hysteresis loop of unaged sample. Moreover, the stabilization of original single domain state during electric field cycling was directly proved by in-situ observation. (See Fig. 1)

This single-domain stabilization is difficult to explain by the domain-wall-pinning effect because there are no domain walls to be pinned in single-domain sample [besides, our recent studies showed aged multidomain Mn-doped BaTiO₃ crystal also has a reversible domain switching (from multi-domain to single domain then back to multidomain). This also cannot be explained by domain wall pinning effect because there would be no domain walls exist in single domain to be dragged back, and pinning force of very short range can not rationalize the restoration of macroscopically displaced domain walls]. Thus the result of our experiment excludes the long-believed mysterious "domain-wall-pinning effect" as the primary origin of aging, and strongly suggests that volume effect is the governing mechanism for ferroelectric aging.

Most importantly, we explained that the volume effect actually comes from a universal symmetry-conforming principle of point defects, which stabilizes existing domain configuration and thus induces aging effect (see Fig. 2). With this clear microscopic mechanism, we are able to explain all known ferroelectric aging effect, including not only the aging in large signal properties (hysteresis loop), but also the small signal aging of dielectric and piezoelectric constant (the hindrance of domain wall motion by defect effect under weak field, as shown in Fig. 2). This study suggests there exists a unified microscopic explanation for all ferroelectric aging.

See L.X. Zhang and X. Ren, Phys. Rev. B. 71:174108, 2005 and 73:094121, 2006 for details.