Using glass content to determine the reactivity of fly ash for thermodynamic calculations

Abstract

Thermodynamic models for the hydration of ordinary portland cement (OPC) typically predict the composition of the resulting pore solution and the hydrates well. However, predictions for cementitious systems containing OPC and supplementary cementitious materials (SCM) are more challenging. The bulk chemical composition of fly ash does not sufficiently reflect the reactive portion of the material, as the crystalline components of fly ash do not generally react in cementitious systems. Thermodynamic modeling inputs using only the bulk chemical composition of fly ash overestimate the extent of both pozzolanic and hydraulic reactions. Two additional approaches are presented to overcome this limitation. In the first approach, the maximum reactive fraction of fly ash is computed by multiplying each bulk phase of the fly ash by a degree of reaction (DoR*) that is measured experimentally through calorimetric methods. In the second approach, the reactive (glass) fraction of the fly ash is determined to calculate its reactivity. In this alternative approach, the fraction of crystalline oxides measured using quantitative x-ray diffraction (QXRD) is subtracted from bulk oxide content determined using x-ray fluorescence (XRF) to establish a degree of reaction for each phase (DoRph*) to be used in the determination of the thermodynamic modeling inputs. Thermodynamic modeling predictions substantially improve by incorporating fly ash reactivity into the calculations using either the DoR* or DoRph*. The calculation of the reactive phases of fly ash using QXRD and XRF data serve as a potential alternative to the current calorimetric methods to calculate reactivity.