ULTRASONIC TRANSDUCER MODELING FOR ACOUSTIC MICROSCOPY & ITS APPLICATION IN BIOLOGICAL MATERIAL CHARACTERIZATION

Abstract

The determination of material properties for very small specimens such as biological cells or semiconductor microchips is extremely difficult and has been a challenging issue for several decades. One important constraint during these measurements is not to harm the specimens during the test process because the specimens, biological cells in particular, are vulnerable to the test itself even during a short period of testing time.Nondestructive evaluation (NDE) is the only suitable precess for such applications. It is fast, causes no disturbance and can give a real time response while being cost effective. Many NDE methods are available today, such as, laser based techniques, Radiography, Magnetic techniques, High resolution photography and other optical techniques, MRI, acoustic and ultrasonic techniques to name a few. Ultrasound is the most popular tool for NDE. As specimens become smaller, the need for shorter wave length ultrasound increases dramatically.The use of acoustic waves in microscopy technology provides many more benefits than its conventional optical microscope counterpart. One such benefit is its ability to inspect a specimen in dark. Another is the capability to see inside an optically opaque specimen. Today, very high frequency, higher than 1 Giga Hertz (109 Hz), ultrasound is being used. This technology has improved at the same pace as the development of electronics and computer science. In acoustic microscopy experiments wave speed and wave attenuation in the specimen are measured by the V(f) technique. A specimen's density, Poisson's ratio and Young's modulus are directly related to the wave speed. V(f) method, as discussed in this dissertation, has some advantages over the more commonly used V(z) method. In order to correctly estimate the wave speed and attenuation in the specimen, the transducer modeling should be completed first. The Distributed Point Source Method (DPSM) is used in this dissertation to model a 1 GHz acoustic microscope lens. Then the model-predicted pressure field is used in a FORTRAN program to calculate the thickness profile and properties of biological cell specimens from experimental data.Transducer modeling at 1 GHz has rarely been attempted earlier because it requires an immense amount of computer time and memory. In this dissertation 1 GHz transducer modeling is conducted by taking advantage of the axisymmetric geometry of the acoustic microscope lens. This exploitation of symmetry in the modeling process has not been attempted prior to this dissertation.