PSI - Issue 5

Adelaide Cerveira et al. / Procedia Structural Integrity 5 (2017) 1116–1122 Author name / Structural Integrity Procedia 00 (2017) 000 – 000

1117

2

leads to l ower binder’s resistance. Concerning the inert/precursor ratio, the nonsignificant differences only occur between the 80/20 and 60/40 cases, although the strength values increase, in general, as the ash percentage increases. © 2017 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the Scientific Committee of ICSI 2017. © 2017 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the Scientific Committee of ICSI 2017

Keywords: alkaline activation; curing conditions; two-way ANOVA; Post hoc tests.

1. Introduction

The use of industrial by-products to produce new types of cement-substitute binders is gaining significant momentum, especially through the alkaline activation technique. However, the exact curing conditions that should be considered with each binder variation have not yet been fully understood. Therefore, this research aims to assess the influence, on the mechanical strength of alkali-activated fly ashes, of different curing conditions (in terms of relative humidity) of several filler/precursor combinations. Mine tailings and fly ash were used as filler and precursor, respectively, while sodium hydroxide was used as activator. Uniaxial compressive strength tests were used to assess the mechanical behaviour of the mixtures, and the data was analysed using statistical tools, Fernández-Jiménez et al. (1999) . Within the framework of this research several mixtures containing fly ashes, sodium hydroxide and sodium silicate were prepared. The fly ash was collected from the Portuguese thermoelectric power plant of Pêgo and, according to the standard ASTM C618-15 (ASTM 2015) was classified as class F. X-ray diffraction (XRD) was used to identify the mineral phases, and quartz and mullite were the main mineral phases detected. The activator used in this work was composed solely by a sodium hydroxide solution, with a specific gravity of 2.13 at 20 ºC (99 wt.%), which was supplied in pellets and later dissolved in distilled and deionised water to obtain a 10.0 molal concentration. The mine tailing was recovered from one of the ponds located inside the mine compound, still in its raw state (i.e. before being subjected to the paste thickener) and transported to the laboratory, where it was separated from most of the constitution water, dried in the oven and, finally, its clods were thoroughly broken. Its chemical composition showed the predominant presence of SO3 (41%), Fe2O3 (28%), SiO2 (18%) and Al2O3 (7%), but also the presence of some heavy metals, like arsenic and lead, although in smaller percentages. Five different mixtures were prepared, with filler/precursor ratios of 80/20, 60/40, 40/60, 20/80 and 0/100. All the specimens were cured at 80ºC for 7 days, after which they were submitted to a uniaxial compression strength (UCS) test. Three different values of curing humidity were considered, namely 25%, 50% and 75%. Each UCS value was the average of 9 different specimens tested. The aim of the present research is to establish how much these two factors (inert/precursor ratio and curing humidity) influence the UCS. To determine the influence of filler/precursor ratio and curing humidity, on the compressive strength a two-way Analysis of Variance (two- way ANOVA) with interaction, was performed followed by a Tuckey’s Post hoc test, when appropriate, Hair et al. (2010), Green and Salking (2008) and Montgomery (2001). In the two-way ANOVA we have three variables: two factors, factor A and factor B, and a dependent continuous variable, X. Each factor divides cases in two or more levels while the dependent variables describes cases on a quantitative dimension. It performs three tests: that the means of the measurement variable are equal for different values of factor A; that the means of the measurement variable are equal for different values of factor B and that there is no interaction between factors A and B. Follow-up tests may be conducted to access specific hypothesis if main effect test, interaction test or both, are significant. The choice of which follow-up procedure to conduct depends on 2. Materials and Methods 3. Statistical analysis