Open Access
Carroll, Travis John
Graduate Program:
Mechanical Engineering
Master of Science
Document Type:
Master Thesis
Date of Defense:
March 16, 2017
Committee Members:
  • Dr. Zoubeida Ounaies, Thesis Advisor
  • Dr. Timothy Simpson, Committee Member
  • PZT
  • fiber
  • piezoelectric
  • interdigitated
  • parallel
  • electrodes
  • electromechanical
  • AFC
Lead zirconate titanate (PZT) is a ceramic material that can be classified as a smart material. This ceramic material exhibits a piezoelectric behavior, which means it is capable of transforming mechanical motion into an electrical charge and vice versa. In order to enhance its use in structural health monitoring systems, PZT, which is typically available in wafer form, is extruded into fibers and used to fabricate active fiber composites (AFCs). In a typical AFC, PZT fibers that are aligned side by side are embedded into an epoxy matrix. Since PZT exhibits both the direct and converse piezoelectric effects, AFCs can be used as actuators, sensors, or both. In order to exploit its piezoelectricity, interdigitated electrodes (IDEs) are applied to the AFC assembly. Due to the special design of the electrodes, the applied electric field is along the fiber’s length, known as the ‘3’ direction. Poling of the fibers is accomplished using these IDEs which results in a poling direction along their length; therefore, the directions of poling and mechanical actuation are the same. This directional manipulation is beneficial because the piezoelectric properties directed along the fiber’s length, or ‘33’ properties, are stronger than those through their thickness, or ‘31’ properties. The piezoelectric response of AFCs is lower than that of bulk PZT due to various factors. Many models have been developed to predict the electromechanical behaviors of AFCs, and many studies have been performed on the mechanical properties of the material components in an AFC. However, very little research has been done on determining the piezoelectric properties of individual PZT fibers in an IDE setup without the presence of the epoxy matrix affecting them. The goal of this thesis is to characterize individual PZT fiber piezoelectric properties in an IDE setup in order to make recommendations for improving the design of AFCs, further optimizing their performance and applications. To accomplish this goal, special attention is focused on how these fibers behave in an IDE setup in order to understand how they would respond electromechanically in an AFC. As previously mentioned, very little research has been done to characterize the piezoelectric properties of individual PZT fibers ex-situ using an IDE setup. To the best of our knowledge, only one research study on electromechanical characterization of individual fibers in an IDE setup has been performed. In this study, done by a former group member, the values obtained through experimental processes and computer-modeling differed. The difference was believed to be caused mostly by various issues with the experimental processes developed at the time. Therefore, this research study extends the previous electromechanical characterization process of individual PZT fibers by improving different aspects of the experimental processes used. The rationale is that improving the experiments will help close the gap between experimentally obtained and model predicted piezoelectric coefficients. The piezoelectric e¬33 and d33 coefficients are the main focus in this study. Experimental values for the induced stress coefficient (e33) are determined for individual fibers in both a parallel electrode and IDE configuration. Various aspects of the experiments are altered in order to improve the reported coefficient values. Additionally, the induced strain coefficient, d33, of individual fibers in an IDE setup is measured. Due to a lack of published results in the literature, finite element analysis (FEA) models are developed in order to validate the induced strain responses of a fiber in both a parallel electrode and IDE setup. Experimentally, e33 values are found to be 4.8 C/m2 in a parallel electrode setup and 3.5 C/m2 in an IDE setup. The IDE coefficient value was about 73% of the parallel electrode value, which was expected from the design characteristics of the IDE setup. Computer models of a fiber in a parallel electrode setup and IDE setup predict the d33 coefficients to be 256 pm/V and 185 pm/V respectively. Again, the IDE value is found to be about 72% of the parallel electrode value. Experimental results report an IDE d33 value of 166 pm/V. The main conclusion of this study confirms that the PZT fiber properties differ from those of bulk PZT; therefore, incorporating these fiber properties into AFC models will lead to more accurate predictions of AFC electromechanical behavior, driving the optimization of AFCs forward.