Alkali-silica reaction: Chemical mechanism, thermodynamic modeling, and effects of lithium ions
The alkali-silica reaction (ASR) is one of the chemical distresses of concrete caused by reaction between reactive silica in aggregates and hydroxyl ions generated by alkalis present in the pore solutions. The present research focused on four areas: (a) chemical sequence and kinetics of the ASR processes, (b) thermodynamic modeling of the chemical sequence of the ASR processes, (c) the kinetics of alkali concentration in cementitious systems, and (d) the exploration of the state of lithium ions in ASR systems.^ The first part of this thesis describes results obtained from the study focused on the kinetics of physical and chemical changes in the reactive aggregate-simulated pore solution system undergoing ASR. Specifically, the study investigated the products formed by exposing reactive silica mineral (α-cristobalite) to the mixture of three alkalis solutions containing solid calcium hydroxide (Ca(OH)2). The experimental results showed existence of a distinct chemical sequence (pattern) of the ASR processes: (a) the formation of calcium silica hydrate, (b) the formation of alkali binding polymerized calcium silica hydrate, c) the increase in concentration of silica ions in the solution, and (d) the formation of ASR gels.^ These experimental results were used to develop thermodynamic model for the chemical sequence and kinetics of the ASR process, including the formulation of the kinetic rate law for silica dissolution (removal of silica from the reactive silica mineral). The innovative features of the kinetic rate law include the ability to account for such factors as pH, temperature, concentration of alkalis in solution, and type of the reactive silica mineral. Subsequently, the proposed kinetic rate low was used as an input to the commercial modeling software (Geochemist's Workbench®) to simulate the chemical sequence of the ASR process. The model generated reasonably accurate predictions of the distribution of species in the reacting system and captured several distinct features of experimental data (i.e. depletion of Ca(OH)2 levels, changes in alkali and silica concentrations, and pH values).^ In addition, the kinetics of concentration of alkali ions in real cementitious system (mortar) undergoing ASR was investigated. The results of these investigations not only confirmed that (as expected) the concentration of alkali ions in pore solution of the system undergoing ASR will decrease but, more importantly, they also revealed that the rate of concentration change is linear with respect to the instantaneous (time dependent) concentration. This represents an important step toward the development of the chemo-mechanical model of ASR as it experimentally confirms the validity of a commonly used assumption of the first order reaction adequately representing the kinetics of the reaction process.^ Finally, the thesis also presents the results of the study on the role of lithium ions in the ASR process. This study was performed using the following three experiments: (a) model reactor experiment (the reactive aggregate-simulated pore solution system undergoing ASR), (b) the analysis of pore solution in mortars containing either reactive aggregate (Jobe sand) or non-reactive aggregate (Ottawa sand) with three different dosages of LiNO3 (0, 0.26 and 0.74 of lithium to molar ratio), and (c) the mortar bar expansion tests performed on specimens exposed to the same three different dosages of LiNO3 as used in experiment (b). The experimental results strongly support one of the previously proposed mechanisms for the role of Li+ ions in controlling ASR. This mechanism involves the formation of the reaction products on the surface of the reactive aggregates, which prevents further hydroxyl ion attack of the particles. This part of the study also revealed that the substantial loss of lithium ions from the pore solution during the hydration period is mainly the result of their incorporation in the newly developing hydration products. In addition, the results of this study were used to develop a model for prediction of the loss of lithium ions from the pore solution during the hydration period. (Abstract shortened by UMI.)^
Jan Olek, Purdue University.