Study of the mechanical stress dependence of hard and soft active ceramic properties is important because many submarine sonar transducers include a compressive mechanical prestress throughout ac electrical activation. The level of prestress to which a ceramic in a transducer is subjected also depends in part on operational factors, such as the level of ac activation and depth of the submersible. This investigation builds upon prior work by Yang et al. by examining the time dependence and the uniaxial stress dependence of the average, differential and dynamic d33 and Y33E of Navy Type III (PZT8) and Navy Type VI (PZT5H) lead zirconate titanate ceramics. This research adds higher levels of prestress and various mid-level ac stress cycles. Under short-circuit conditions, large and small compressive stresses are applied to the samples while measuring dielectric displacement and strain. The piezoelectric coefficient, d33, is evaluated using the direct method as a function of time, prestress level, and ac stress magnitude. The constant-field modulus is calculated from the slope of the corresponding stress-strain curves. Intrinsic and extrinsic contributions to these properties are discussed.
It is a well-known fact that electrostrictive materials, such as lead magnesium niobate-lead titanate (PMN-PT) ceramics, exhibit significant frequency dispersion in their small signal dielectric constant below their dielectric maximum temperature Tm. The frequency dispersion in several PMN-PT compositions will be examined in this study using two independent measurement methods: dc biased resonance and large signal quasistatic measurements conducted on NUWC Division Newport's SDECS. From these measurements, the coupling factor, piezoelectric constant and Young's modulus are compared as a function of the applied bias and frequency. Both the DC biased and SDECS measurements were performed on the same 3:1 aspect ratio samples. Finite element calculations will show that the error in determining the Young's modulus and piezoelectric constant from resonance using these samples is less than 5 percent. It will be shown that when frequency dispersion exists it remains even with the application of dc bias, and that the degree of deviation between these quantities increases the further below Tm the temperature drops. It will also be shown that, like the dielectric constant, the coupling factor, piezoelectric constant and Young's modulus in PMN-PT ceramics above Tm are non-dispersive.
The large signal performance of electrostrictive materials, such as lead magnesium niobate-lead titanate (PMN-PT), is of critical importance to sonar transducer and actuator designers. However, obtaining these large signal parameters properly, particularly under compressive prestress, is an expensive and time-consuming enterprise. The complexity of these measurements, therefore, precludes them as a method for quickly and easily screening materials for their potential as high power materials. Traditionally, resonance measurements, which otherwise are relatively simple to perform, have been used for screening purposes, but they suffer from the drawback that the material parameters obtained are at the incorrect frequency and under no prestress. Furthermore, it was unclear what significance the results of resonance measurements for nonlinear materials such as electrostrictors had. It has recently been suggested that dc biased resonance measurements on electrostrictive ceramics would be an accurate predictor of the coupling factor and optimum bias point. In this paper, dc biased resonance measurements on three different PMN-PT formulations, with varying dielectric maximum temperatures, will be analyzed to determine which composition has the highest predicted coupling factor. This prediction will be compared with large signal quasistatic measurements conducted on NAVSEA Division Newport's SDECS (Stress-dependent Electromechanical Characterization System). The predictive ability of the resonance measurements will also be analyzed as a function of temperature normalized with respect to the dielectric maximum temperature.
The dielectric loss factor tan (delta) is a critical parameter in transducer design and performance prediction, as it is directly related to the electrical energy lost to Joule heating. A method for calculating the equivalent loss factor of 'quasi linear' materials from a measured major polarization vs. field loop by extending the standard definition of tan (delta) for a linear lossy capacitor to nonlinear materials is presented. To extract effective loss tangents for minor loops from the major loops, an area correction algorithm was implemented. This algorithm proves to be nearly exact for all bias and drive levels in the case of an ideal linear capacitor. In most cases, the effective tan (delta) calculated from a major loop agrees fairly well with that calculated from a directly measured minor loop. Finally, the behavior of the loss tangent as a function of the dc bias field, ac drive field, prestress level and temperature will be examined. It shall be shown that, in general, the effective loss factor of lead magnesium niobate-lead titanate decreases with increasing temperature, consistent with its transition from a piezoelectric to an electrostrictive material, but increases with prestress. At a fixed temperature and prestress level, the loss factor increases as the bias or drive levels decrease. However, at certain bias levels, the loss tangent is practically the same, regardless of the drive level.