Fingerprinting and assessing the reactivity of past and future aggregates

Abstract

The Irish construction boom of the early 2000s generated an unprecedented demand for sub floor rock aggregate. However, due to a lack of residential specific material standardisation, unacceptable documentation practices by aggregate retailers and the by passing of any geological input by property developers, deleterious pyritic rock aggregate was emplaced within the foundations of up to 12, 500 homes across Ireland. For almost a decade after the construction of these houses, debate has raged within the Irish and European court system in order to find and bring to account the quarry sources responsible for supplying such unsuitable hardcore material. This thesis investigates the use of pyrite geochemistry alongside advanced machine learning models, to develop an identification mechanism for such pyritic aggregate quarry sources. Within this technique, pyrite major element concentrations are used alongside pyrite trace element concentrations, whole rock δ34S and total sulphur concentrations to comprehensively characterise each quarry source investigated. In addition to this, the geochemical variation of pyrite within an Irish aggregate quarry source was also investigated to ensure that the geochemical variation of the analysed samples was representative of the variation seen within a quarry source. This research gives confidence that, a quarry source may be determined for an unknown pyritic rock aggregate sample taken from a pyritic heave effected house. This thesis also attempted to investigate the relationship between pyrite trace element concentration and oxidation rate. Unfortunately, this investigation proved unsuccessful due to an unforeseen decrease in SO4 % following an artificial oxidation experiment. Nevertheless, by investigating a correlation between these two variables it is hoped that a much improved experimental procedure can be carried out in the future to more definitively investigate the effect of pyrite trace element concentration of oxidation rate.