TY - CONF T1 - Modelling Piezocomposites with Varying Porosity and Interconnectivity A1 - Dent,A.C.E. A1 - Lewis,R.W.C. A1 - Bowen,C.R. Y1 - 2009/// KW - Piezoelectric KW - Composite KW - 3-3 Connectivity KW - Microstructure KW - Micromechanical model KW - Finite Element Modelling KW - Porous KW - Design SP - 187 EP - 187 JF - Program of MRS 2007 Fall Meeting T2 - MRS 2007 Fall Meeting CY - Warrendale PA, USA PB - Materials Research Society N2 - For certain applications the use of high density piezoceramics may not be desirable, such as sensing low frequency hydrostatic waves for sonar. Research into this field has established that piezoelectric composites with a low-density and high porosity can provide improved hydrostatic properties and high figures of merit. Piezocomposites with interconnecting struts in three-dimensions (3-3 connectivity), are well suited to hydrostatic applications with simple manufacturing routes via foam replication or burning out sacrificial polymer spheres. However, producing such materials with extensive porosity can result in regions that are isolated mechanically and / or electrically, such that a range of connectivity patterns from 3-3 to 0-3 may occur locally. To determine how these complex microstructures may be optimized, it is necessary to develop micromechanical models that can represent the piezocomposite with varying degrees of porosity and interconnectivity. Micromechanical models for 3-3 piezocomposites have relied on an idealised unit cell containing a single pore. This oversimplification has contributed to the observed differences between experimental and model results. With increased computational power and efficient finite element modelling techniques, large models have been created with numerous randomly placed pores, filled with a passive secondary phase at a predefined volume fraction. The resulting geometry features a variety of pore types and distributions depending on the volume fraction of porosity, from isolated to fully interconnected pores. This modelling approach better represents porous microstructures and has provided numerical predictions in good agreement with experimental data. Through modelling large ensembles of stochastic microstructures, the variability in composite properties is also revealed, as is found in practise. The modelling approach developed was applied to investigate key parameters including pore distribution, aspect ratio and volume fraction. The predicted composite properties (d33, d31, K33, dh and gh) were used to calculate relevant hydrostatic figures of merit. Based on the simulated microstructures and performance indices, processing and material properties were identified for improved hydrophone materials. Such models are valuable for guiding piezocomposite design, when selecting a compromise between maximising a particular figure of merit and minimising the potential variation of performance between samples and devices. Y2 - 2007/11/26/ M2 - Boston MA, USA. ER -