Setting and nanostructure of slag-fly ash binders

Slag fly ash binders (SFB) are room temperature hardening binders that have high strength, reasonable setting time, and lower carbon footprint when compared to ordinary Portland cement (OPC). The precursors for SFB are industrial byproducts, namely slag and fly ash. The inexpensive nature of the precursors offer SFB a cost-competitive edge over OPC and makes them an attractive alternative binder. Calcium silicate hydrate type gels (C-S-H, C-A-S-H, and C-N-A-S-H) coexist with geopolymers in SFB. While C-S-H is composed of discrete silicate chains, geopolymers are made of a three-dimensional aluminosilicate framework. The desirable properties of SFB coupled with the feasibility of large-scale production have prompted many researchers to study this rather complex system. However, little is known about the setting and nanostructure of these binders as a function of key variables like curing time, curing temperature, and slag/fly ash ratio. This knowledge is vital to create standards and building codes when these binders are used large-scale. The aim of this work was three-fold. The first aim was to characterize the precursors. The precursors were systematically characterized using x-ray diffraction (XRD), energy dispersive x-ray fluorescence (EDXRF), magic angle spinning - nuclear magnetic resonance spectroscopy (MAS-NMR), Fourier transform infrared spectroscopy (FTIR), and laser diffraction. These characterization techniques highlighted the variability of the precursors. A small amount of penta-coordinated Al (Al(V)) was found in slag. The information gained from the abovementioned methods would serve as a guide for future researchers to make an informed choice while selecting precursors for binders. The proportion of unreacted fly ash was found to be high in SFB. Hence, fly ash was replaced with metakaolin because the latter is more reactive than the former. Since there were many commercial metakaolins available in the market, five of them were characterized using the methods mentioned above. Out of those five metakaolins, two were chosen to make slag metakaolin binders (SMKB). The Al environments of commercial metakaolins were different from one another which influenced the reactivity of metakaolin in the binders. The second aim was to understand setting in SFB using two complementary methods, namely shear wave ultrasonic wave reflectometry (SUWR) and modified ASTM C403 penetration test method. The disparity in set time measured by these two tests suggested that a “soft” gel which had a solid percolating network, but no gel strength existed in SFB. This soft gel later gained gel strength. Higher the slag/fly ash ratio, larger was the difference in initial and final set times measured by the two methods. So the structure of the soft gel in the SFB with the highest slag/fly ash ratio was analyzed. MAS-NMR results revealed that the soft gel was composed of a nascent C-N-S-H gel and orthosilicate (Q0) units. The amount of C-N-S-H gel formed was low when compared to later ages, and it had no Al incorporated in its gel structure. The proportion of orthosilicate (Q0) units was high when compared to later ages. The third aim was to identify and quantify the poorly ordered phases as a function of curing time, curing temperature, and slag/fly ash ratio. This goal was achieved by using selective chemical extractions and MAS-NMR spectroscopy of binders and extraction residues. C-N-A-S-H gel was found to be the major binding phase and geopolymer the minor binding phase. The mean chain length (MCL) of C-N-A-S-H gel had a negative correlation with slag/fly ash ratio and positive correlation with curing temperature. Although the proportion of geopolymers increased in SMKB, still C-N-A-S-H gel was found to be the dominant product phase. Therefore, the availability of Al(V) does not necessarily favor the formation of geopolymer over C-N-A-S-H gel in SMKB. The amount of unreacted metakaolin in SMKB was found to heavily depend on the proportion of Al(V) in the precursors. The reason for this is because Al(V) was the most reactive species in metakaolin followed by Al(IV) and then Al(VI).U of I OnlyAuthor requested U of Illinois access only (OA after 2yrs) in Vireo ETD syste