Date of Completion


Embargo Period



George A. Rossetti, Jr.; Ramamurthy Ramprasad

Field of Study

Materials Science and Engineering


Master of Science

Open Access

Open Access


The domain structure of ferroelectric and multiferroic materials can have a significant effect on piezoelectric, dielectric, and thermal transport properties. Piezo Force Microscopy is an ideal tool based on Atomic Force Microscopy that allows unique investigations of such nanoscale effects, and can further be implemented to monitor domain switching dynamics.

A new method for polarization orientation mapping and statistical analysis is first employed to determine the domain variants present in a range of BiFeO3 epitaxial (001) thin films with specifically engineered domain distributions. This allows domain wall densities to also be calculated, along with interfacial polarization angles between adjacent domains (ferroelectric 180°, and ferroelectric and/or ferroelastic 109° and 71° interfaces). Domain walls can be identified as charged or un-charged as well, which interestingly is identified for the first time as depending on the horizontal or vertical alignment of the domain boundary. For certain domain engineered specimens, particularly those with only 2 domain variants present, this leads to charged interfaces exclusively along and neutral interfaces along , therefore providing a route for unique, direction dependent future ferroelectric or multiferroic devices. Furthermore, increased domain wall densities are shown for the first time to inversely correlate with thermal conductivity, suggesting that domain walls scatter phonons similar to grain boundaries. Again, this can be used to engineer unique transport properties for future ferroelectric and multiferroic devices.

The domain polarization process itself is also investigated using PFM. For epitaxial (001) PbZrTiO3, movies of consecutive domain maps are acquired during the switching process itself. Analysis of domain wall positions as a function of poling time therefore reveals domain growth velocities, which are determined in a variety of directions. Results are presented based on a range of prepoled domain patterns, designed to isolate domain wall velocities as a function of crystallographic directions as well as possible AFM scanning artifacts. Experimental artifact effects are in fact negated, with domain growth enhanced perpendicular to the AFM fast scanning axis regardless of the crystallographic alignment. Initial domain patterning conditions are found to influence domain growth, however, likely suggesting charge depletion or accumulation in the PZT film adjacent to pre-poled structures. Such insight is crucial for ferroelectric domain engineering efforts and the ultimate performance of ferroelectric devices.

Major Advisor

Bryan D. Huey