Recent Advances in Characterizing the Interfacial Mechanical Properties of Composites based on Ultrasonic Inversion

Abstract

In response to the growing demand for composite materials with high specific strength, high specific stiffness, and lightweight structural characteristics, this paper systematically reviews the development and research progress of ultrasonic characterization techniques for fiber–matrix interfaces. The mechanical and durability properties of composites are largely governed by the interfacial bonding quality and its evolution mechanisms. However, interfacial damage is often concealed within the material, highlighting the urgent need for quantitative and nondestructive monitoring approaches. First, the paper reviews the emergence and development of the Scanning Acoustic Microscope in the 1970s and elaborates on its pioneering role in revealing interfacial microstructures and early-stage debonding phenomena. Subsequently, it focuses on the quantitative studies since the 1990s based on the imperfect interface spring model, including the introduction of normal and tangential stiffness parameters (KN and KT), the development of reflection/transmission inversion methods, and the establishment of spectrum–angular spectrum analysis techniques. These advances enabled a paradigm shift in interfacial characterization—from mere visualization to quantitative evaluation. In the 21st century, research has further evolved toward refined modeling, optimized inversion algorithms, and engineering-oriented applications, leading to the formation of multi-angle, multi-frequency, and multimodal integrated detection frameworks. By integrating signal processing, finite element simulation, and machine learning, the accuracy and stability of interface parameter inversion have been significantly enhanced. Finally, the paper summarizes the major challenges in this field, including model ill-posedness, adaptability under complex environments, and difficulties in data fusion. It also envisions future developments toward intelligent, real-time, and multi-physics-coupled characterization approaches.