Geopolymer from a Western Australian fly ash

Abstract

Ordinary Portland cement is utilised worldwide as a mainstay construction material. Worldwide consumption of cement in 2009 was estimated to be 2.8 billion tonnes, which unfortunately equates to the production of 2.8 billion tonnes of CO[subscript]2 via the sintering procedure required to produce cement clinkers. With worldwide concern over climate change, this value is a substantial contribution to greenhouse gas emission.Fly ash is a by-product of coal combustion from thermoelectric power stations. World production of fly ash was estimated at 600 million tonnes with only 9% utilisation (Kayali 2007). The remainder is typically disposed of in landfills or ash ponds. This takes up usable land for development, introduces environmental hazards and can lead to undesirable events with an example being the rupture of a fly ash pond barrier at the Kingston Fossil Plant, Tennessee in 2008 (Reilly 2008) known as the TVA Spill.Geopolymer is a cementitious binder and is considered an environmentally friendly alternative to cement as it emits no CO[subscript]2 during production. Production of geopolymer is a simple process that involves mixing an amorphous aluminosilicate feedstock with alkaline activating solution. In addition this process is able to utilise low cost industrial by-products such as blast furnace slag and fly ash as feedstock.Although fly ash is suitable as a feedstock for synthesis of geopolymer its inherent heterogeneity limits development of a general formulation for processing geopolymer. Beneficiation of fly ash can be considered a method for alleviating this limitation, leading to a more homogeneous geopolymer with improved properties.Collie fly ash from Western Australia was selected as the fly ash to investigate as it is the dominant fly ash in the State and had been successfully used previously to make geopolymer. The amorphous content of Collie fly ash was determined by dissolution and a combination of QXRD and XRF. Collie fly ash was thoroughly characterised by QXRD and XRF, revealing a reactive amorphous content of 54.5 wt.% and secondary phases of carbon, hematite, maghemite, magnetite, mullite and quartz. The amorphous component was found to contain a modest amorphous iron oxide (5.5 wt.%) which after dissolution studies and subsequent analysis by QEMSCAN, was determined not to play a direct role in geopolymerisation. Crystalline quartz was found to exist as primary quartz separate from the fly ash spheres and secondary quartz embedded in the spheres believed to have exsolved from the decomposition of clay in the production of mullite.Beneficiation of the fly ash was conducted in a three stage procedure using sieving, milling and magnetic separation to improve fly ash homogeneity and reactivity. Sieving was effective in reducing large carbon and free primary quartz content. Interestingly most of the carbon was found to be small and finely dispersed throughout the material making it unfeasible to remove by sieving. Sieving in conjunction with milling increased surface area from 9.83 m[superscript]2/g to 10.7 m[superscript]2/g. Magnetic separation revealed that the amorphous iron was not magnetic though the complete removal of crystalline iron phases is not possible without a robust separation technique. The removal of magnetic phases increased the surface area of the sieved and milled fly ash to 12.9 m[superscript]2/g.At each stage of beneficiation the proportion of reactive amorphous material increases resulting in increased reactivity. This increase in reactivity necessitated changes in solids:liquids ratio to maintain a workable geopolymer mixture. The least beneficiated fly ash (sieved < 45 μm) produced the strongest geopolymer with a compressive strength of 132 MPa. The most beneficiated fly ash (sieved/milled/magnetically separated) produced geopolymer with the same compressive strength as geopolymer from unmodified fly ash (100 MPa). However, the highly beneficiated fly ash geopolymer proved to be highly resistant to high temperature cracking even after exposure to 900 ºC. The outcomes from this project clearly identifies that different levels of fly ash beneficiation lead to different geopolymer properties which in turn extend the range of applications for which geopolymers can be used.