| AVS 55th International Symposium & Exhibition | |
| Surface Science | Wednesday Sessions |
| Session SS2+NC-WeM |
| Session: | Functional Metal Oxides and Quantum Metal Structures |
| Presenter: | P. Maksymovych, Oak Ridge National Laboratory |
| Authors: | P. Maksymovych, Oak Ridge National Laboratory S. Jesse, Oak Ridge National Laboratory M. Huijben, University of California, Berkeley R. Ramesh, University of California, Berkeley A. Morozovska, National Academy of Science of Ukraine S. Choudhury, Pennsylvania State University L.-Q. Chen, Pennsylvania State University A.P. Baddorf, Oak Ridge National Laboratory S.V. Kalinin, Oak Ridge National Laboratory |
| Correspondent: | Click to Email |
Nucleation of domains is central to understanding and applying ferroelectric materials. In bulk ferroelectrics, the number of ferroelectric nuclei grows rapidly with temperature, verifying the involvement of thermal fluctuations, which are widely interpreted as a result of defect, modified local barriers. In homogeneous, low defect thin films, a local voltage source can produce intrinsic switching domain without the influence of defects. We have examined the temperature depended of intrinsic domain nucleation for the first time using Piezoresponse Force spectroscopy of model oxide materials, BiFeO3 and Pb(Zr0.2Ti0.8)O3 from 34 to 300 K. Measurements were performed in ultra high vacuum, which allowed cryogenic cooling but also precluded the presence of water known to significantly affect ferroelectric phase stability. The key finding is that the temperature dependence of ferroelectric switching initiated by a scanning probe tip is small and amounts to only a 20 percent increase in the nucleation voltage upon cooling of the films to the lowest temperature. These results are in stark contrast to previous reports for ferroelectric switching in the capacitor geometry where the coercive field increases by at least a factor of five over a similar temperature range. To complement the experimental data, a theoretical analysis was carried out based on two models, the rigid ferroelectric model and phase-field simulations. The temperature dependence predicted by the phase-field modeling is small and in good agreement with the experiments, while the rigid ferroelectric model overestimates it considerably. We conclude that the intrinsic ferroelectric switching induced by scanning probe experiments involves minimal contribution from thermal fluctuations. PM was supported as a Eugene P. Wigner Fellow at ORNL. Research was performed at the Center for Nanophase Materials Sciences, Oak Ridge National Laboratory, and supported by the Office of Basic Energy Sciences, U.S. Department of Energy.