Nanoscale studies of domain wall motion in epitaxial ferroelectric thin films

Abstract

Atomic force microscopy was used to investigate ferroelectric switching and nanoscale domain dynamics in epitaxial Pb͑Zr 0.2 Ti 0.8 ͒O 3 thin films. Measurements of the writing time dependence of domain size reveal a two-step process in which nucleation is followed by radial domain growth. During this growth, the domain wall velocity exhibits a v ϰ exp− ͑1/E͒ dependence on the electric field, characteristic of a creep process. The domain wall motion was analyzed both in the context of stochastic nucleation in a periodic potential as well as the canonical creep motion of an elastic manifold in a disorder potential. The dimensionality of the films suggests that disorder is at the origin of the observed domain wall creep. To investigate the effects of changing the disorder in the films, defects were introduced during crystal growth ͑a-axis inclusions͒ or by heavy ion irradiation, producing films with planar or columnar defects, respectively. The presence of these defects was found to significantly decrease the creep exponent , from 0.62-0.69 to 0.38-0.5 in the irradiated films and 0.19-0.31 in the films containing a-axis inclusions.

Nanoscale studies of ferroelectric domain walls as pinned elastic interfaces

“…Artificial linear domains are thus "written" in the sample by a scanning tip held at a voltage beyond the switching threshold, while individual nanodomains are created by the application of voltage pulses to a stationary tip. Such nanodomains, whose size depends on both the duration and magnitude of the voltage pulse, as well as the size of the tip itself [77,78], can be as small as a few nm in radius [79], in particular when ultrasharp tips based on carbon nanotubes are used [80,81,82], and remain fully stable in measurements extending over a year [83,84]. While PFM imaging is clearly neither real-time, nor full-field-of-view, information on the switching dynamics and domain growth rates in the sample can nonetheless be obtained from averaging the size of sufficient numbers of these nanodomains written with a particular voltage pulse duration and magnitude.…”

“…Independently, following magnetooptical studies in ferromagnetic materials [97], detailed analysis of the roughness and growth of individual, artificially-written domain walls in epitaxial Pb(Zr 0.2 Ti 0.8 )O 3 thin films by Paruch et al, Tybell et al [85,98,24] demonstrated non-linear creep motion at low fields, with a dynamical exponent µ = 0.5-0.6, and power-law growth of the relative displacement correlation function B(r) ∼ r 2ζ at small length scales below the ferroelectric sample thickness, with a roughness exponent ζ = 0.25, as can be seen in Fig. 5(a).…”

“…Rather, to tackle the effects of random fluctuations at multiple length scales, a statistical description is necessary. Such a statistical, highly reductionist approach to domain walls within the general framework of disordered elastic systems has proven very powerful, allowing metastability and the glassy physics associated with an energy landscape characterised by multiple local minima to be addressed [24,25,26]. Conversely, as a result of the high resolution of scanned probe microscopy, combined with access to a broad range of control parameters such as the applied electric field, temperature, environmental (electrochemical) and strain boundary conditions, ferroelectric domain walls present a useful model system for testing theoretical predictions about pinned elastic interfaces.…”

Ferroelectric or non-ferroelectric: Why so many materials exhibit “ferroelectricity” on the nanoscale

“…From the applied perspective, this interest derives from their current use in a wide variety of active electromechanical and electrooptical elements in devices such as transducers, actuators, motors, sensors, non-volatile memories, etc. [2][3][4][5][6][7] Similarly, from the physics viewpoint, they provide a system to explore a gamut of functionalities ranging from the motion of elastic interfaces through disordered media and associated scaling laws for domain wall motion, [8][9][10] possibility of localized metal-insulator transitions [11][12][13][14] and order parameter coupling at ferroelectric domain walls, [15][16][17][18] statistical modeling of macroscopic polarization switching by microscopic switching units, [19][20][21][22][23] unique behaviors under spatial confinement, 24,25 and many more. Thus, the study of ferroelectrics has remained pertinent even 90+ years since their initial discovery in a Rochelle salt, in 1921 26 .…”

Domain pinning near a single-grain boundary in tetragonal and rhombohedral lead zirconate titanate films

“…17,18 Experimentally, PFM measurements demonstrated that 1-and 2-dimensional defects produce significant pinning of 180º domain walls in ferroelectric films. [19][20][21][22][23][24] The local electric field created by defects and domain wall -domain wall pinning are the major contributions to domain wall pinning. 17,25 It is widely reported that in tetragonal ferroelectrics the domain structures of adjacent grains are correlated due to local strain and electric fields.…”