Crack Self-Healing in Alkali-Activated Materials: From Mechanisms to Strategies for Enhancing the Healing Efficiency
Abstract
The increasing concern for decarbonizing concrete manufacturing is calling for sustainable construction materials since portland cement production contributes about 8% of global anthropogenic carbon dioxide emissions. Alkali-activated materials (AAMs) have emerged as promising alternative binders to curb carbon dioxide emissions from portland cement production and allow more sustainable construction. In addition, AAMs have desirable mechanical properties and high resistance to acidic and sulphate attacks.
Like conventional concrete, AAMs are quasi-brittle materials with low tensile strength, demonstrating susceptibility to cracking. Cracking-initiated deteriorations such as reinforcing steel corrosion, sulphate attack, and freeze-thaw damage, pose severe threats to the service life of concrete structures. Yet, sustainability is not readily achievable even if structures were made of environmentally friendly AAMs. Increasing awareness of sustainability has promoted the development of crack self-healing technologies. While crack self-healing in cement-based materials has been a topic of extensive research, very few studies have so far investigated the self-healing of AAMs. Therefore, there is need to explore self-healing approaches for improving the crack self-healing capability of AAMs.
This dissertation is centered on exploring the possibilities of enhancing the self-healing efficiency in fibre-reinforced alkali-activated slag-based composites via using calcium hydroxide powder, crystalline additives, expansive minerals, and biomineralization. The self-healing effect was evaluated using a portfolio of techniques, including optical microscopy, compressive and tensile tests, sorptivity tests, mercury intrusion porosimetry (MIP), inductively coupled plasma optical emission spectroscopy (ICP-OES), and X-ray microcomputed tomography (μCT). Crack self-healing products were characterized using scanning electron microscopy with energy dispersive X-ray (SEM-EDS) analysis and Raman spectroscopy. Ultimately, a chemistry-informed machine learning model was proposed to estimate the compressive strength of AAMs, therefore guiding the design of AAMs that meet the requirements for construction. The results demonstrate that all healing strategies explored herein improved the crack closure ratio significantly. In addition, the mechanical properties, watertightness, and pore structure were enhanced. SEM-EDS revealed that the main self- iii healing product was calcium carbonate. The computational intelligence model developed permits accurate prediction of the mechanical strength of AMMs and the effects of influential mixture design parameters. The findings of this study should bridge the knowledge gaps in crack self-healing of AAMs, thereby removing the hindrances that impede broader implementation of AAMs in construction applications.