Impact of single crystal properties on underwater transducer designs

Abstract

The ongoing robotic revolution in oceanic science puts new requirements on underwater transducers. Small platforms require compact multi-purpose transducers, and there is limited space available for the electronics. Introducing single crystal ferroelectrics as the active material can be one way of meeting the new requirements. Compared to lead zirconate titanate (PZT) polycrystalline ceramics, single crystals provide higher electromechanical coupling, higher strain per electric field and higher elastic compliance. A higher electromechanical coupling coefficient, k, enables an increase in frequency bandwidth. With more frequencies available, more tasks can be covered using only one type of transducer, saving space on the small platforms. Higher elastic compliance reduces the thickness of the transducer element, and larger strain per electric field enables reduction of the voltage source. The object of this Ph.D.-project was to investigate the impact of the single crystal properties on underwater transducer designs, keeping in mind the current development of textured ceramics promised to provide single crystal-like properties at a lower cost. For many underwater transmitters, the usable frequency band is restricted by both the transmitted acoustic power and the reactive electrical power. Single crystals have the potential to double the usable band compared to PZT, but the acoustic matching required for this can be difficult to obtain in practice. We investigated this for a 1-3 piezocomposite plate and for a tonpilz. The composite plate had air backing for high efficiency. Tonpilz transducers are common for resonance frequencies below 50 kHz, while piezocomposites are widely used for higher frequencies. We started by exploring simple 1D models for a plate, using real, frequency independent acoustic loading, and we observed that an increase in k has a larger bandwidth impact on the electrical power factor than on the acoustic power. For piezocomposite plates, acoustic matching to water is achieved by adding matching layers. These matching layers are frequency dependent, and when optimized for a maximally flat passband, the matching decreases rapidly outside the passband. Poor acoustic matching prevents the bandwidth of the electrical power factor from reaching its potential. For many underwater applications, the diversity provided by a selection of narrow-band pulses at a variety of frequencies is more important than a flat acoustic response. We proposed to extend the passband by separating the resonances of the complex system constituted by the piezocomposite plate and the matching layers. Using this strategy, we modelled and fabricated a single crystal 1-3 piezocomposite transducer with air backing and two matching layers, achieving reactive power below 50% in a frequency band 175 % wide relative to the resonance frequency 518 kHz. The piezocomposite had matrix material Epotek 301-2. The fabricated composite had an effective coupling coefficient of k = 0.83, in good agreement with the modelled result. Reported single crystal underwater transducers are mainly of the tonpilz or cylinder design. The herein presented successful fabrication of a piezocomposite transducer that can be used in the frequency range 244 kHz to 1148 kHz shows that single crystals are indeed interesting also in the high end of the frequency range applied for underwater applications. The tonpilz and cylinder design does however have an advantage in their inherent frequency independent matching to water. Our modelling showed that a single crystal tonpilz transducer with k = 0.82 can be designed with a mechanical quality factor as large as 2 and still exhibit reactive power below 20% in a frequency band of width 150 % or below 50 % in a band of width 170%, relative to the resonance frequency. Single crystals provide high coupling also in modes for which the electric field and the main extension are in separate directions. We investigated the 32 mode utilized in a 1-3 single crystal piezocomposite, and presented a design for which the acoustic power at a given voltage was estimated to increase by almost a factor 50 compared to a conventional PZT piezocomposite that utilizes the 33 mode. An additional benefit of the 32 mode design was almost 50 % reduction in composite thickness. We also presented a design for which a mode akin to the 31 mode was included to add an extra, usable, frequency band to a 32 mode transducer. It was concluded that the 32 mode design opens for transducers that can be operated over a wide frequency range and driven by low voltages, making it well suited for mounting on compact platforms.