Gutierrez Page 1 Piezoelectric Anisotropy in (1 x )NBT x BT Single Crystals for 0.0 9 < x < 0.16 Michael J. Gutierrez 1 2 1 Department of Materials Science and Engineering, University of Florida, Gainesville, FL 32611, USA 2 School of Materials Science and Engineering, University of New South Wales, NSW 2052, Australia ABSTRACT Piezoelectric materials have many applications in modern electronics including power generation and energy harvesting, sensors and transducers, as well as actuators. (1 x )NBT x BT has been gainin g considerable attention as an alternative to PZT, one of the most commonly used piezoelectrics. In order to more fully understand this material, coercive field and d 33 measurements were undertaken These measurements were obtained on single crystals of co mpositions approaching the 6.5% barium titanate MPB of (1 x ) NBT x BT normal to the (001) and (110) crystal planes These measurements show that (1 x ) NBT x BT exhibits increasing anisotropy and extender behavior as the amount of barium titanate in the system is increased above 6.5% barium titanate composition
Gutierrez Page 2 I. INTRODUCTION In recent years, ferroelectric 100( (1 x )NBT x BT ) has been gaining attention as a possible environmentally friendly alternative to the piezoelectric lead containing PZT 1 Its ferroelectric and piezoelectric properties are due to its non centrosymmetric tetragonal perovskite crystal structure in which the central titanium is offset above the central position Due to this, an applied electric field will cause the material to strain, known as the converse piezoelectric effect. Equally an applied force on the material will produce a charge, known as the direct piezoelectric effect. This phenomenon is described by the piezoelectric coe fficient, d 33 which denotes that a field applied i n the z direction, will produce a strain in the z direction. Often the ferroelectric and piezoelectric properties of a material are enhanced at the morphotropic phase boundary in which the structure of the material changes. This is due to the additive response of multiple ferroelectric phases. A piezoelectric material can be described as either an extender or a rotator depending on its dominant strain direction in response to an applied electric field. Oft en the magnitudes of the d 33 and d 15 are used to define the dominant strain direction Here d 33 refers to the extent that an applied electric field in the 3 direction induces a strain in the 3 direction, while d 15 refers to the extent that an applied elect ric field in 1 direction induces a shear strain about the 2 direction as seen in Figure 1 where: [11 12 13 [1 6 5 12 22 23 = 6 2 4 13 23 33] 5 4 3] If the dominant piezoelectric response is found along the polar axis th e n d 33 along the polar axis is greater than d 15 and the material is known as an extender. Conversely, in rotator piezoelectrics the maximum response is due to polarization rotation and d 15 in the perpendicular
Gutierrez Page 3 direction is greater than d 33 in the polar direction. 4 However, the d 15 can be difficult to measure. It is possible to indirectly assess the d 15 by cutting the crystal such that the (110) face is exposed. The d 33 as measured normal to this crystal plane correlates with the d 15 This method will be used to obtain all measurements d 15 measurements. In order to more fully characterize this material, ferroelectric and piezoelectric properties of single crystals must be measured Previous studies have often placed p articular interest o n the behavior of this material near its morphotropic phase boundary (MPB). However, the behavior of piezoelectric and ferroelectric properties approaching the MPB can also be insightful. For example t he pi ezoelectric behavior of barium titanate has been shown to switch from acting predominantly as an extender to a rotator along other MPBs 2 Since (1 x )NBT x BT exhibits a ferroelectric tetragonal to ferroelectric rhombohedral MPB at 6.5% barium titanate as seen in Figure 2, compositions approaching this phase boundary will be examined 3 II. EXPERIMENTAL Polarization and strain hysteresis loops were measured using a TF Analyzer. The crystals were first cut using a high precision diamond saw such that the ( 001) or (110) face was exposed. Sample dimensions ranged from approximately .25 mm 3 1 mm 3 Each sample was then coated with silver paint electrode on its ( 001 ) or (110) top and bottom faces and polished to remove all traces of electrode on side faces. In the TF Analyzer, samples were subjected to increasing electric fields ranging from 2500 6000 kV/mm at 0.1 Hz. Afterwards samples were cleaned using a acetone bath in an ultrasonicator for 2 minutes to remove all silver paint electrode. Samples were then annealed at 450 C for 10 minutes and allowed to cool on the hotplate for 30 minutes in order to remove all remnant domains. This process was then repeated for additional tests For each composition, the coercive field was identified as the average absol ute field as
Gutierrez Page 4 the polarization hysteresis loops crossed the electric field axis. The d 33 coefficients were calculated as the slope of linear portion of the strain hysteresis loops seen in red in Figure 1 III. RESULTS AND DISCUSSION In comparing the differences between the polarization and strain hysteresis loops, the effect of composition and direction on the ferroelectric and piezoelectric properties of NBT xBT can be determined. Figure 1 clearly shows this effect. As the percentage of barium titan ate increases, not only does the magnitude of the strain but the piezoelectric coefficient increases as well. This result is expected as barium titanate is highly piezoelectric. It is can also be seen that the (001) direction demonstrates a higher strain a nd piezoelectric coefficient compared to (110) direction. This too is expected as strain in the (001) direction is preferential due to the offset titanium ion. Figure 2 also demonstrates that an increase in the percentage of barium titanate in the system results in a decrease in the coercive field as well. These m easurements can be seen in Table 1. Of particular interest is the increase in the anisotropy of the piezoelectric coefficients between the (001) and (110) directions. As seen in Figure 3, the perc entage of barium titanate has a much greater effect on the piezoelectric coefficient in the (001) direction than it does in the (110) direction. While the d 33 for the (001) direction is less than 2 times greater than that in the (110) direction at 9 percen t barium titanate, it is over 6 times greater at 16 percent barium titanate. At this composition, NBT xBT can be classified as an extender piezoelectric because it strains much more parallel to the electric field than perpendicular to it. However, because the anisotropy between these directions decreases as the MPB is approached, a possible transition from an extender piezoelectric behavior to a rotator piezoelectric behavior is observed.
Gutierrez Page 5 I V. CONCLUSIONS As expected for NBT xBT the d 33 increases and the coercive field (E c ) decreases with increasing barium titanate composition: the coercive field lowers and the piezoelectric coefficient increases. Additionally, NBT xBT possesses higher d 33 values parallel to the  direction relative to the  direction More importantly the anisotropy between these two directions increases more dramatically with increasing barium titanate concentration. Overall, NBT xBT exhibit s an extender like beh avior. As th e concentration of barium titanate decreases and the MPB is approached, this anisotropy greatly decreases and there is a n incipient transition from overall extender to rotator like behavior in this material. Further characterization of the ferr oelectric and piezoelectric behavior of NBT xBT at and below the MPB is desired. V. Tables and Figures Property Crystal Cut 9BT 12BT 16BT E c (kV/mm) (001) 2.27 2.12 1.45 (110) 4.19 1.38 1.97 d 33 (pm/V) (001) 219 858 1765 (110) 153 191 284 d 33 Anisotropy Ratio (001) (110) 1.43 4.49 6.21 Table 1. Coercive fields and piezoelectric coefficients for different compositions and directions of (1 x)NBT xBT.
Gutierrez Page 6 Figure 1 Strain as a function of electric fields for different compositions and directions of NBT xBT. Figure 2. A morphotropic phase boundary (MPB) exists at roughtly 6.5% barium titanate for (1 x)NBT xBT. 5
Gutierrez Page 7 Figure 3 Strain as a function of electric fields for different compositions and directions of NBT xBT.
Gutierrez Page 8 Figure 4 Polarization as a function of electric fields for different compositions and directions of NBT xBT. Figure 5 Measurement s of d 33 vs the composition of (1 x)NBT xBT
Gutierrez Page 9 REFERENCES  J.Daniels et al ., Appl. Phys. Lett. 98 252904 (2011)  D. Damjanovic et al. Appl. Phys. Lett. 80 652 (2002)  T. Takenaka, K. Maruyama, K. Sakata, Appl. Phys. Lett. 30 2236 2239 (1991)  M. Dais, M. Budimir, D. Damjanovic, N. Seter  Jones, Aksel et al. Sensors, Volume 10, Issue 3 ACKNOWLEDGMENTS National Science Foundation under award number OISE 1129412 (IRES Grant) Dr. Jacob Jones Dr. John Daniels: UNSW Supervisor Richard Quio: UNSW Grad Mentor