Effect of alkaline activator composition on the geopolymer properties produced from South African power station fly ash

Abstract

Eskom produce ~36 million tons of ash annually, of which 90 % is fly ash. Only 7 % of the ash is valorised; much lower than the global average of ~75 %. Geopolymer production from fly ash, as a Portland cement substitute, presents an opportunity to upcycle this underutilised resource. Geopolymers require significantly lower energy inputs compared to cement production and on average emit 0.45 kg of CO2 per kg produced whereas cement production emits 1 kg of CO2 per kg produced. To date there has been limited research on geopolymers using South African fly ash as a feed stock. Specifically, the formulations of alkaline activator (a sodium silicate and sodium hydroxide mixture) required to produce geopolymers of the required strength and durability at competitive cost needs to be determined. The overall aim is to develop a cost-effective geopolymer formulation using South African power station fly ash as a housing building material. The significance of this investigation is that it could valorise an underutilised waste stream from coal combustion, with significant reductions in energy and greenhouse gas emissions compared to Portland cement. Phase one determined the sodium silicate to sodium hydroxide ratio required to produce the target geopolymer strength (30 MPa) and crystalline properties. The variables studied were sodium hydroxide concentrations, curing age (3, 7 and 28 days) and curing temperature (21 and 80 ºC). The ultimate compressive strength was found to increase by 32 % with sodium hydroxide concentrations increasing from 10 M to 15 M. A sodium silicate to sodium hydroxide ratio of 2.5 performed notably better than a ratio of 1.5. Curing at the elevated temperature significantly increased geopolymer strength compared to curing at ambient temperature. The optimum geopolymer formulation was made using 15 M sodium hydroxide, the highest sodium silicate to sodium hydroxide ratio (2.5) and cured at 80 ⁰C for 6 hours and achieved a compressive strength of 23 MPa. XRF, XRD, TGA and FTIR analyses were used to characterise the geopolymers. The XRD results show a decrease in amorphous phase content from 60% in the fly ash to an average of 52 % in the geopolymers. This is primarily caused by the decrease in mullite as it is the aluminosilicate source that is decomposed by the sodium hydroxide. The Si: Al ratio, calculated from the XRF results reveal that the geopolymer with the highest compressive strength, has the highest ratio while the geopolymer with the lowest strength has the lowest

ratio. The TGA showed an 8 – 10% weight loss between 0 and ~200ºC associated primarily with water loss. Drying shrinkage, expansion on rewetting and soundness are required tests according to SANS 1215 for concrete masonry units. The drying shrinkage results average 600 microstrain which is out of specification by a factor of 10 and is most likely caused by a high water content. Expansion on rewetting and soundness test results, however, were well within specification. An economic evaluation was conducted to compare with cement brick prices in industry. The geopolymer cost calculations were based on operating cost which included cost of the materials used and the cost of electricity for oven curing. The geopolymers cost between R1.50 and R1.79 per brick while cement bricks cost R2.00 per brick on average. This shows that the geopolymers are cost effective when compared to cement bricks.