Mechanisms of improved transport phenomena in mature portland cement concrete pavements: A macro and microstructural evaluation.

Abstract

The objective of this research is to improve the understanding of fluid transport through portland cement concretes and cement pastes under both unsaturated and saturated conditions. The sorptivity property is of primary interest for the unsaturated condition, and permeability and chloride ion penetration resistance (measured by rapid chloride permeability) are considered under saturated conditions. For unsaturated conditions, the key factors influencing sorptivity are water to cement ratio, degree of hydration, sample age, effective porosity, degree of sample saturation, sample drying temperature, and aggregate volume fraction. Porosity measurements from the water absorption test are in agreement with predictions by Powers and Brownyard's classical model when the sample moisture state is accounted for. Using statistical analysis, a four parameter multi-variate linear model that includes water to cement ratio, degree of hydration, degree of sample saturation, and aggregate volume fraction is proposed to predict sorptivity. The model allows the concrete designer to estimate sorptivity in concrete based on mix design, hydration level, and moisture state. Under saturated conditions, experimental permeability measurements are in rough agreement with existing microstructure-based models when using pore structure measurements made directly on concrete samples, with emphasis given to models by Dullien (a combination of the Hagen-Poiseuille and Darcy equations), Carman and Kozeny (a hydraulic radius model), and Katz and Thompson (a formation factor model). However, making microstructure measurements on concretes instead of cement pastes does not improve model accuracy. Permeability and rapid chloride permeability measurements for ordinary portland cement concretes are well correlated to compressive strength, and are significantly impacted by climate zone and distance from the exposed concrete surface. Rapid chloride permeability results are influenced by both the physical pore structure and the conductivity of the solution in the pores. Supplementary cementitious materials reduce calcium hydroxide content and in turn pore solution conductivity and rapid chloride permeability measurements. Changes in calcium hydroxide content are used to predict changes in the conductivity of the pore solution and deviations from the expected relation to porosity. This allows the concrete designer to forecast rapid chloride permeability based on water to cement ratio, hydration level, and chemical composition.