STUDIES ON ITZ AND DAMAGE CHARACTERISTICS IN CONCRETE OF VARYING STRENGTHS: AN ANALYTICAL AND EXPERIMENTAL APPROACH

Abstract

The interfacial transition zone (ITZ) plays a vital role in determining the mechanical as well as fracture properties of concrete, making it a topic of significant importance in the field of concrete technology. The interfacial transition zone refers to the region surrounding the aggregate particles in concrete, where a distinct transition occurs between the aggregate surface and the cement paste matrix. It is characterized by variations in composition, porosity, and bonding strength, which significantly influence the overall performance of the concrete. Understanding the characteristics of the interfacial transition zone is crucial for gaining a comprehensive understanding of the mechanical behaviour and fracture performance of concrete structures. The transition zone plays a significant role in facilitating efficient stress transfer between the aggregates and the surrounding cement matrix, thereby ensuring the overall structural integrity of the material. Any weaknesses or defects present within the ITZ can hinder the efficient mechanism of load transfer, resulting reduction in both the strength and stiffness of the material. Furthermore, the fracture properties of concrete are significantly influenced by the inherent properties of the interfacial transition zone. Firstly, owing to its distinctive attributes compared to the bulk cement matrix, the ITZ serves as a focal point for stress concentration, rendering it increasingly prone to the initiation and propagation of cracks. Secondly, variations in the properties of the ITZ, including thickness, porosity, and bond strength, also contribute towards cracking behaviour, deformation characteristics, and the ultimate failure of concrete structures. In view of the importance of the ITZ in concrete performance, this PhD thesis aims to comprehensively examine the characteristics of the ITZ in various strengths of concrete, including normal strength concrete, high strength concrete, and ultra-high-performance concrete. The study employs a combination of experimental, analytical, and machine learning approaches to gain a deeper understanding of the ITZ and its influence on the mechanical and fracture properties of concrete. Firstly, the work focuses on the characterization of the interfacial transition zone by evaluating the ITZ thickness and volume fraction. Cement phase RVE has been simulated, and ITZ thickiii ness has been evaluated whilst considering the hydration of cement particles. The limitations associated with the assumption of uniform transition zone thickness around the aggregate have been overcome by considering differential ITZ, which provides a realistic representation of the interfacial phase. The concept of Voronoi Tessellation has been employed to quantify the volume fraction of the ITZ in concrete containing spherical and polygonal aggregates, assigning differential ITZ thickness around the aggregates. A more accurate and realistic representation of the transition zone in concrete is realized by utilizing this method. The obtained results were carefully validated against existing experimental data available in relevant literature sources, thus ensuring the reliability and accuracy of the proposed analytical framework for ITZ characterization. Further, experimental investigations have been performed to study the fracture processes in pre-notched beam specimens with varying strengths under normal and high-strength categories. The acoustic emission (AE) technique has been utilized during the experimentation for detecting and analysing microcracks and their propagation within the concrete matrix. By studying the fracture behaviour, this objective seeks to provide insights into the mechanical response and damage mechanisms of different strengths of concrete, shedding light on the role of the interfacial transition zone in these processes. Acoustic emission waveform results have been utilized for the damage source classification in different grades/strengths of concrete through a machine learning approach. Additionally, artificial neural network (ANN) based models have been developed to differentiate between different concrete strengths (say, normal strength and high strength), aiming to establish a correlation between damage sources and concrete strength. The sequential feature selection technique has been utilized to identify the most significant features that contribute to strength discrimination in normal and high-strength concrete. Further, the proposed machine learning-based classification algorithms have been extended to classify various damage stages of ultra-high-performance concrete (UHPC). By training a model using AE waveform data belonging to different damage stages (elastic region, strain hardening, plateau and strain softening) in UHPC, a robust and accurate classification system has been developed that can assist in assessing the structural health and condition of ultra-high-performance concrete under different stress and loading conditions. Implementing machine learning in this context opens up new possibilities for optimizing concrete mix designs and improving quality control procedures.