An ultrasonic methodology for determining the mechanical and geometrical properties of a thin layer using a deconvolution technique

An ultrasonic method is proposed for simultaneously determining the thickness, density, sound velocity, and attenuation of a thin layer from a reflection spectrum at normal incidence. The normal theoretical reflection spectrum of a thin layer is established as a function of three dimensionless parameters to reduce the number of independent parameters. The inverse algorithm, using the least squares method, is adopted to determine the dimensionless parameters, and the corresponding convergence zones are investigated. The measured reflection spectrum at normal incidence is obtained using Wiener filtering, and spectral extrapolations following Wiener filtering are applied to obtain the time-of-flights by identifying the overlapping pulse-echoes inside the thin layer and the superposing pulse-echoes from the reference material and front surface of the specimen. The thickness of the thin layer can then be calculated and as initial estimate for the inverse algorithm. The density, sound velocity, and attenuation are then determined by the measured thin layer thickness and determined dimensionless parameters. Two 500 mu m stainless steel and aluminum plates were immersed in coupling water and a 5 MHz flat transducer was applied. The relative errors of measured thickness, density, and sound velocity were less than 6%, and the ultrasound attenuation was close to its true value. The validity of the proposed technique was verified. (C) 2013 Elsevier B. V. All rights reserved.