Chloride ingress into concrete has long been known to decrease the service life of built infrastructure. Inadequate knowledge of the physical reasons associated with chloride diffusion into concrete could generate chloride penetration profiles that become meaningless for prediction of service life. In this study, the effects of pore closure (physical effect) and changes in chloride binding (chemical effect) on chloride diffusion through Australian General Purpose (GP) cement pastes were investigated. Through-diffusion tests and "in-and-out" diffusion tests were conducted to monitor the time-dependent chloride diffusion through cement pastes cured from 1 to 28 days. The through-diffusion test quantified the overall chloride diffusion behaviour at different stages of cement hydration, which was a combined result of physical and chemical processes controlling diffusion. The "in-and-out" test differentiated the contributions of the physical and chemical process on the chloride diffusion at different stages of cement hydration. As expected, the reduction of chloride diffusivity was significant during the first two weeks of curing, most likely attributed to the significant reduction of porosity as well as establishment of capillary discontinuities within the pore structure. It was also observed that the amount of bound chloride was not constant but increased significantly from 1 to 28 days of curing age.
