A Theoretical Study On Chloride Transport In Concrete Subjected To Freeze-thaw Cycles

Existing concrete structures in cool coastal areas are prone to be attacked by both chloride attack and freeze-thaw cycles(FTCs).The combined deterioration process accelerates the damage evolution of concrete and reduces the service life of concrete structures.Based on the mechanism of chloride attack and freeze-thaw cycles,the time-varying porosity is adopted in both chloride binding isotherm and ionic diffusion coefficient to consider the pore evolution induced by FTCs.Therefore,the time-varying porosity can not only reflect how freeze-thaw action affects the pore structure of concrete,but also couple freeze-thaw process together with ionic diffusion process at time scale.This study presents a numerical study on the mechanism of chloride transportation under freeze-thaw cycles.Firstly,considering the random distribution of aggregates,a series of two-dimensional and three-phases model,including mortar,aggregates and ITZs,are established.Secondly,by considering ionic binding effect,multi-ion electrochemical coupling,temperature effect,etc.,the time-dependent relationship between the concentration of bound and free chlorides as well as the spatial-temporal distributions of multi-species is obtained from this model.Afterwards,through the detailed modelling of multi-phase and multi-species,a systemic parametric analysis is carefully performed on a series of significant influencing factors,i.e.,water-to-cement ratio,hydroxyl concentration,external chloride concentration and temperature variation,which affect chloride binding and transport remarkably during freeze-thaw cycles.This paper processes the results of numerical models using Grey Relational Analysis and Bayesian Updating Method.Grey Relational Analysis is conducted to rank the correlative degrees of influencing factors on diffusivity of chloride under FTCs,which can bring insights to the preparation technique of concrete in cold coastal regions.Bayesian Sense is applied to combine the common regularity of numerical models and individual information of experiments by updating the penetration depth,which can calculate a more reasonable penetration depth to predict the durability of concrete structures.