INVESTIGATING THE GEOCHEMICAL IMPACT OF SO2 CO-INJECTION IN GEOLOGIC CO2 SEQUESTRATION: AN EXPERIMENTAL AND MODELLING APPROACH

Captured emissions from coal fired power plants are likely to contain small amounts of SO₂. The geological co-sequestration of SO₂ is expected to intensify brine acidification resulting in enhanced mineral dissolution and precipitation. Experiments on SO₂-rock-brine reactions were conducted and converted into geochemical models to improve our understanding and modelling capability on CO₂-SO₂ storage. Synthetic brine was acidified with H₂SO₄, an aqueous proxy of SO₂, and reacted with sedimentary rock samples from the Western Canada Sedimentary Basin, Canada, and the Surat Basin, Australia. Different sample particle sizes, starting pH and temperature were applied to investigate their impact on reaction rates and to enhance the ability to upscale to reservoir conditions. The experimental outcome indicated that the rate and extent of H₂SO₄ induced reactions strongly depend on mineralogy. A low pH, higher temperature and/or smaller particle size increased the observed reaction rates. The experimental data was integrated into kinetically controlled reaction path models to identify and quantify mineral dissolution and precipitation. In all cases the modelling provided a good fit to the experiments indicating that the reaction rate parameters derived from published sources were applicable. The modelling of the experiments with different particle size required increases in reactive surface area with decreasing particle size for the reactive silicate minerals including feldspar, muscovite and chlorite. However, in order to model the dissolution behaviour of the carbonate minerals, the reactive surface area had to be decreased with increasing particle size, consistent with expectations for pore filling authigenic mineral phases. This has significant implications for the modelling at reservoir scale. The outcomes of reaction path modelling at reservoir scale of these rocks with CO₂-SO₂ mixtures can thus be impacted depending on how upscaling is conducted, particularly with respect to the extent of dissolution and precipitation over the short term (0 to 100 years). In a final step the models will be extended to longer time frames to investigate long term effects and integrated into reactive transport models of CO₂ injection to investigate the impact of CO₂-SO₂ storage on different temporal and spatial scales.