Reactivity of crystalline slag phases in cementitious systems

Portland cement manufacturing is thought to account for as much as 4-8% of global CO2 emissions. Efforts to incorporate industrial wastes into concrete in order to offset the environmental burden of traditional cements have been growing in recent years. Reactive industrial wastes such as blast furnace slag and fly ash have been the subject of much research and have been successfully implemented in blended cements as partial replacements for Portland cement. Other wastes, such as steel and copper slags, have been largely neglected due to their lower reactivity. Concurrent with efforts at incorporating industrial wastes into concrete has been the development of alternative concretes with different chemistries, e.g. alkali-activated materials, blended Portland cements, and calcium sulfo-aluminate cements. This thesis focuses on improving our understanding of how the constituent phases of steel and copper slags interact with the aqueous phase of concrete, with the goal of identifying suitable applications in concrete. The focus on the constituent phases is recognition of the fact that steel and copper slags are too variable in composition to study on a case-by-case basis, and only through studies of their constituent phases can we identify opportunities for their use. First, I quantify the effect of aqueous chemical environment on the dissolution rate of the minerals calcio-olivine and fayalite, identified as primary phases of ladle furnace steel slag and copper slag, respectively. Calcio-olivine and fayalite were exposed to solutions of NaOH and Ca(OH)2, chosen to mimic the pore solutions of alkali-activated and Portland-cement based binders. These results have significance for incorporating high olivine content steel and copper slags in concrete. Second, I elucidate the effect of aggregate surface chemistry on the type, morphology, and rate of reaction product formation in Portland cement-type systems. Polished surfaces of limestone, quartz, fayalite, and diopside were exposed to Ca and Si rich solutions and the resultant reaction products were characterized using scanning electron microscopy. The results of this study indicate that calcium silicate hydrate (the dominant reaction product of Portland cement concretes) nucleation and growth kinetics are accelerated on limestone surfaces relative to quartz, fayalite, and diopside surfaces, although no differences in the morphology of the precipitated C-S-H is observed. The relevance of this research is in the importance of the type and kinetics of reaction product formation on the surface of an aggregate, which plays a crucial role in the development of hardened, load-bearing concrete. This experimental research is supplemented by extensive literature review of the composition and microstructure of steel and copper slags, as well as the dissolution rates and thermodynamics of dissolution in concrete pore solution of the relevant steel and copper slag phases. This work serves to contextualize the previous experimental research and to contribute towards the development of a kinetic model that accounts for the reaction kinetics of both Portland cement and crystalline slags (steel or copper slag). This thesis also presents a methodology for calibrating pH meters in highly alkaline solutions such as those relevant to cementitious systemsPh.D