Optimization of matching layer design for medical ultrasonic transducer

This thesis work contains two major parts. In the first part, ultrasonic wave propagation in multilayer structure is investigated. Delaminations between ceramic and electrode layers in multilayer capacitors and multilayer actuators are common defects, which are difficult to detect using traditional ultrasonic imaging method if the size is smaller than 50 microns in diameter. The T-Matrix method is used to treat wave attenuation in periodic structures with alternating ceramic and electrode layers. Multiple penny-shaped delaminations are assumed perpendicular to the incidence wave, and the forward scattering amplitude of the wave from delaminations is calculated by substituting the average effective crack opening displacement into the scattered wave displacement. The effective phase velocity, wave amplitude and the attenuation coefficient have been calculated for different crack densities. The results provide a theoretical base for potential attenuation based ultrasonic non-destructive evaluation (NDE) method.;The second part is a study on matching layers. Matching layers are crucial components in ultrasonic transducers for medical imaging. Without proper matching layers, large acoustic impedance mismatch between piezoelectric resonator and the human body tissue will cause most of the ultrasound energy to be reflected at the interface. For a given frequency, the matching layer thickness should be one quarter of the wavelength and its acoustic impedance should be the geometric mean of the acoustic impedances of piezoelectric material and the imaging body. There are no natural materials that can precisely meet such requirements. Therefore, solid particle/polymer composites are commonly used as matching layer materials. The acoustic impedance of such composites is generally in the range of 2-15 MRayls. It is a routine task to make such composite for low frequency transducers, but for transducers with operating frequency higher than 40 MHz, the powder size must be sub-microns in order to reduce wave scattering because the wavelength is much smaller. High volume loading fine powder composite is very difficult to make using conventional composite fabrication technique because air bubbles will be trapped in the mixture. Therefore, all ultrahigh frequency transducers currently used or under development are not properly matched because the lacking of desired matching layer materials. This problem hinders the development of finer resolution ultrahigh frequency ultrasonic imaging.;There are some progress made in the past 3 years and there are sol-gel SiO2/polymer nano-composites being developed that can have acoustic impedance up to 5.7 MRays. In this thesis work, TiO2 nano-structured material has been developed. The material has porous nano-structure with the volume fraction of voids being controlled by the amount of bonding amorphous phase in the material and its acoustic impedance can be further tuned by heat treatment at slightly elevated temperatures. Using the quarter wavelength thickness characterization method, the acoustic properties of this nanostructured material were accurately characterized. It was found that the acoustic impedance can reach as high as 7.19 MRayls, which is a concrete improvement compared to that of the best nano-composites available.;Because of recent rapid development of the single crystal PMN-PT and PZT-PT materials for medical ultrasonic transducer applications, there is a new excitement to develop transducers with very broad bandwidth because the electromechanical coupling coefficient of these single crystals are better than 90%. For such very broad bandwidth transducers, the front matching layer will be the limiting factor because the quarter wavelength matching layer acts like a filter whose bandwidth is generally less than 100%. One of the main tasks of this thesis is to investigate matching layer design with gradient acoustic impedance to achieve much broader bandwidth (>100%). Wave propagation within an inhomogeneous multilayer structure has been analyzed and simulated using the finite difference time domain numerical technique. By adjusting the acoustic impedance distribution function, we have found the best gradient design which has super broad bandwidth. In fact, the passband only has a low frequency cut-off and it works for almost all frequencies beyond the cut-off frequency. To certain extend, this optimized design is universal so that such a matching layer can be used in all medical transducers of different frequencies.