GB's effects in polycrystalline ferroelectric thin films
-
1. From SS-PFM, do grain boundaries seem to have great influence on the local switching behavior? Compared to point defects and dislocation, which one has greater effects?
2. Usually there are multi-domains inside one grain, is there a way to get a single domain within a grain? Is PFM a good way to do this, or are there some better ways?
Thanks -
Grain boundaries very definitely have an effect on switching in the sense that they provide local variations in lattice structure which a domain wall must negotiate to grow past it. They can also provide nucleation sites.
The smaller the grain, the more likely to get single domains. -
GBs affect swithcing behavior very strongly by serving as nucleation centers for noew domains, but also as pinning centers for domain wall motion. The degree of influence obviously depends on type of GB (clean vs. with precipitates, orentation angles).
On bicrystals, it is possible to explore the mechnisms of domain nucleations at GBs systematically:
BRIAN J. RODRIGUEZ, SAMRAT CHOUDHURY, Y.H. CHU, ABHISHEK BHATTACHARYYA, STEPHEN JESSE, KATYAYANI SEAL, ARTHUR P. BADDORF, R. RAMESH, LONG-QINQ CHEN, and SERGEI V. KALININ, Spatially Resolved Polarization Switching at an Engineered Defect: Bicrystal Grain Boundary in Bismuth Ferrite, Adv. Func. Mat. 19, 1 (2009)
The pinning was studied by several groups, including Ramesh (papers with Ganpule), Gruverman (I do not have reference at hand, but he is also on this forum), Huey (not sure if published), and Balke (not sure if published). Also, see
BRIAN J. RODRIGUEZ, Y.H. CHU R. RAMESH, and SERGEI V. KALININ, Ferroelectric domain wall pinning at a bicrystal grain boundary in bismuth ferrite, Appl. Phys. Lett. 93, 142901 (2008).
and some papers by N. Valanoor. -
One of the first papers demosntrating domain pinning on grain boundaries and their effect on the polarization relaxation after poling in PZT films:
V. V. Shvartsman, A. L. Kholkin, Investigation of switching behavior in PbZr0.45Ti0.55O3 thin films by means of Scanning Probe Microscopy, Ferroelectrics 286, 291 (2003). -
Also, Nina Balke at ORNl is doing some spectacular work. Contact her at n2b@ornl.gov
-
Great! thanks for the answers and references!
If GBs serve as nucleation center, it seems to me that it would be easier to switch the area close to the GBs. If GBs serve as pinning centers, it would be harder to switch the area close to the GBs. If this is true, what determines the role of GBs? whether it's nucleation center or pinning center.
How does the orientation angles come into play? -
Both. It can be a nucleation and a pinning centers. The mechnisms of nucleation and wall motion are fairly different.
There is a very good book by Sidorkin that addresses some of htese issues.
Orientation angle comes into play by determining what happens at GB. If it is tilt GB that is formed by parallel dislocation cores, and will affect poalrization weakly since th elattice "almost" continuous. If the GBs are generally misoriented, the supression of poalrization must be much stornger since reocnstructions have to be much more extensive. -
There are some great reviews by S. Pennycook and G. Duscher that address structures of tilt GBs in perovskites. Caveat is that electron microscopists (unsurprisingly) prefer GBs where they can get atomic resolutions on both constituent crystals, and this means special orientations.
-
The grain boundary can induce a local electric field that affects polarization close to the boundary. Gruverman has a nice paper demonstrating this with chemical deposition there. And we know this from theory as well... see Rappe's papers on BTO.
The amount of charge will depend on structure, i.e. orientation. This has been measured in non ferroelctric perovskites and will occur in ferroelectrics as well.
See Shao et al APL 2004, PRL 2005
and a nice paper by the Stuttgart group is also referenced in the PRL mentioned above. -
The role of GB in switching will depend on type of defects it is associated with. Large-angle GB with high dislocation density in its vicinity would pin polarization. Accumulation of vacancies would increase GB conductivity and may make it easier to switch provided that it does not cause too much leakage.
-
Our recent paper in APL 93, 222905 (2008) directly proves the existence of frozen dipole moments at the grain boundaries in pure SrTiO3 ceramics. These frozen dipoles are responsible for many peculiar phenomena such as forbidden Raman lines,
stiffening of soft mode and reduction of the permittivity as compared to single crystals. -
Actually, conductivity impedes switching, since field becomes smaller