Characterization of Ferroelectric BaTiO 3 (100) Surfaces by Variable Temperature Scanning Surface Potential Microscopy a
- PDF / 1,559,658 Bytes
- 6 Pages / 417.6 x 639 pts Page_size
- 89 Downloads / 139 Views
Mat. Res. Soc. Symp. Proc. Vol. 596 © 2000 Materials Research Society
heating stage. To perform piezoresponse measurements, our AFM was additionally equipped with a Wavetek function generator and SRS830 lock-in amplifier. W 2C coated tips (1 , 125 rtm, resonant frequency - 350 kHz) were used for these measurements. During variable temperature experiments, temperature was increased in steps of -10'C and the system was kept at the selected temperature for -0.5 h in order to achieve thermal equilibrium. In SSPM measurements, the cantilever was re-tuned at each step. Thermal drift was corrected by adjusting lateral offsets with respect to domain-unrelated topographical features. Further details of measurements and image processing are provided in [14-15]. A BaTiO 3 single crystal (5x5xl mm, T,--130 0 C, Superconductive Components, Inc) was used for these studies. The characteristic roughness of the (100) face did not exceed 15 A. Prior to analysis the crystal was repeatedly washed in acetone and deionized water. In order to obtain a reproducible well-developed domain structure the crystal was heated above Curie temperature, kept at 140'C for -0.5 h and cooled down on a metallic heater surface. RESULTS Surface topography and surface potential of a region of the BaTiO 3 surface at various temperatures are shown on Fig. 1. Characteristic surface corrugations aligned in the y-direction 0.62' 9 indicate the presence of 900 a-c domain boundaries. The corrugation angle 0 determined by AFM is in excellent agreement with theoretical value 0= n/2-2arctan(a/c), where a and c are the parameters of the tetragonal unit cell (0--0.6290). A number of small spots due to contaminants are also present. Surface potential reveals 1800 domain walls separating c-domains of opposite polarity aligned preferentially in x-direction. Potential variations across 900 a-c and 1800 c÷-c domain walls results in characteristic "checkerboard" pattern. Similar domain structure is reported in [16]. The dark spot indicated by the arrow represents surface contamination that significantly (- 70 mV) depresses the surface potential. On increasing the temperature, the contaminants and the local potential depression were used to adjust for thermal drift. The overall domain structure (i.e. the number and relative size of domains) doesn't change on heating from room temperature to temperatures just below T,; however, the surface corrugation angle which is directly related to c/a ratio in the tetragonal unit cell, changes with temperature. In order to obtain a reliable measurement of corrugation angle, it is averaged over the y-direction and over 4 domains for each image. Fig. 2a. shows the typical corrugation profile and average corrugation angle as a function of temperature. The solid line represents a calculated corrugation angle based on the temperature dependent values of a and c for tetragonal BaTiO 3 extracted from reference [17]. The Curie temperatures for the crystal under investigation (130'C) and that studied by Kay and Vousden (120'C) were matched by
Data Loading...
