HIGH STRENGTH RUBBERIZED PERVIOUS CONCRETE FOR SUSTAINABLE PAVEMENTS: ENGINEERING PROPERTIES AND LIFE CYCLE ASSESSMENT

Abstract

This study investigates the performance of Pervious Concrete (PC) with varying quantities of Natural Coarse Aggregate (NCA) replaced by Waste Tire Rubber (WTR) whereas Supplementary Cementitious Materials (SCMs) such as Silica Fume (SF) and Fly Ash (FA) were employed as substitute for cement. The considered percentage of WTR were 5%, 10% and 15% by weight of NCA. Cement was partly substituted by FA (10% constant) and SF (5%, 10% and 15%). Physical and mechanical performance of pervious concrete were examined using permeability indices (density, porosity, and water permeability) and strength indices (compressive, flexural, and splitting tensile strengths, along with pervious concrete deterioration). X-Ray Diffraction (XRD) and Scanning Electron Microscopy (SEM) were conducted to determine the microstructural variations, hydration products precipitation and Interfacial Transition Zone (ITZ) development. Porosity and permeability were increased by utilizing WTR. while causing a reduction in the mechanical properties of PC. However, the addition of SF and FA enhanced the strength indices of pervious concrete due to the post-hydration pozzolanic reactivity and gap filling effect whereas, leading to a decrease in the permeability. Developed constitutive models were implemented to compare the stress-strain response of permeable concrete with experimental stress-strain curves. Life Cycle Assessment (LCA) showed that WTR, SF, and FA-based PC lessened the greenhouse gas (GHG) emissions of pervious concrete and could potentially be utilized as an environmentally benign substitute, hence promoting the circular economy. Furthermore, this study presented a machine learning-based predictive modeling to refine the mix design of PC in order to improve structural performance without compromising permeability. The experimental validation produced a high coefficient of determination and low error values, demonstrating that machine learning is effective in optimizing the mix proportion of pervious concrete.