Characterization Of Curved Samples Using Laser Photothermal Infrared Radiometry Technique

Photothermal Radiometry (PTR) is a new detection technique based on photoacoustic (PA) technology. It has become a very powerful tool for the thermophysical characterization and the nondestructive evaluation (NDE) of a wide variety of materials because of its high sensitivity and nondestructive nature. For decades, researches in photothermal techniques have been restricted to samples with flat surfaces. With the increasing applications of PTR to the characterization of materials with cured surface, studies on curved surface samples (e.g. cylindrical or spherical samples) have been motivated in recent years. In this thesis, first, we develop a method to study the characteristics of inhomogeneous cylindrical and spherical using the quadrupole method, in which cylindrical samples are studied both theoretically and experimently. Second, the relationship between an inhomogeneous cylindrical or spherical sample and an inhomogeneous flat sample is studied theoretically. A signal processing method is developed to eliminate or partially eliminate the effect of surface curvature on PTR signals for the first time, in which the PTR signals of a homogeneous and an inhomogeneous curved sample with the same structure are normalized. It is found theoretically that the curvature effect of both cylindrical and spherical samples can be eliminated or partially eliminated under certain conditions, the normalized PTR signal from the curved inhomogeneous sample can be approximately equivalent to that of a flat surface with the same structure.