| dc.description.abstract | In this research, a coupled chemical reactive transport model is developed for the simulation of calcite dissolution under CO2 storage conditions. The coupled model is applied to systematically evaluate the relative effects of pore-structure parameters and their evolution during calcite dissolution. Pore-
structures are represented using different pore networks which vary in pore-size distribution and average pore connectivity (i.e. coordination number). The model discretises the 3D pore-network generated using PoreFlow software into several cells over which PHREEQC performs 1D reactive transport calculations. Through averaging, PoreFlow outputs the pore-throat radii and total specificreactive surface area for each cell, which is needed for the reaction calculations using PHREEQC. Subsequently, PHREEQC outputs the calculated amount of dissolved calcite for PoreFlow where it is redistributed over the network according to the relative residence times of the various pore-throats. For simplicity it is assumed that dissolution only occurs within pore-throats and that pores with a shorter residence time dissolve more than pores with a longer residence time. Through this coupling, reactive transport calculations can be done at a relatively low computational cost whilst still taking into account the evolution of pore-structure. Results of reactive transport simulationsindicated a relatively high reactivity which can be attributed to the high temperature and pressure conditions but could also be caused by an overestimation of the rate considering the system may not be well-mixed for the velocity used. Notably, the average coordination number is found to have a negligible effect on the evolution of the pore-throat radius distribution. Furthermore, networks with a small mean pore-size are found to be much more sensitive to dissolution and show more drastic variations for changes in standard deviation. This may be attributed to small pores increasing at a much faster rate than large pores. Moreover, because pore networks with a smaller mean pore-size have a higher density of pores they will also have relatively more reactive surface area and will thus dissolve faster than networks with a higher mean pore-size. The relationship between residence time and pore-throat radius is shown to be insufficient to explain the observed evolution of pore-throat distribution and thus gives further evidence of the complexity of the system. | |
