This paper presents an experimental study of carbon dioxide (CO2) mineralization by injecting aqueous nanobubble dispersion of CO2 into crushed basaltic rock samples at 75°C. Two CO2 nanobubble fluids were compared: one with deionized (DI) water and the other with 30 wt% sodium formate solution (20 wt% formate). Each injection was initialized by saturating the rock sample container with DI water without CO2 content. The main objectives of this research were to test two mechanisms of mineral dissolution, proton-promoted and ligand-promoted mechanisms, and their effects on the subsequent CO2 mineralization with metal silicates in basaltic rock samples.
The experimental program in this research included measurements of CO2 contents in aqueous nanobubble dispersions, static experiments of mineral dissolution by sodium formate solutions at different concentrations, and dynamic flow-through experiments of mineral dissolution and CO2 mineralization by the two injection fluids mentioned previously.
Aqueous nanobubble dispersion enabled the water phase to be saturated with CO2 at a higher level than the inherent solubility of CO2 at 75°C. At 138.9 bara, the CO2 concentration in the nanobubble fluid with DI water was 1.75 mol/L, which was 65% greater than the inherent solubility at the same temperature and pressure. The dynamic flow-through experiments with CO2 nanobubble dispersion in DI water (Case #1) resulted in an enhanced level of mineral dissolution, where the highest concentrations of Mg and Ca ions were 88.1 and 63 ppm, respectively. The metal ion concentrations decreased with time likely because of the formation of passivating layers. SEM analysis of the rock samples did not indicate in-situ CO2 mineralization in Case #1.
The other dynamic experiment with CO2 nanobubble dispersion in 30.2 wt% sodium formate solution (Case #2) confirmed an even greater level of mineral dissolution than Case #1. The highest concentrations were 407 and 504 ppm for Mg and Ca ions, respectively. Unlike Case #1, the mineral dissolution did not diminish with increasing throughput of the injection fluid in Case #2, although the CO2 concentration in the fluid in Case #2 (1.09 mol/L) was smaller than that in Case #1 (1.75 mol/L). SEM analysis of the basaltic rock samples after the experiment in Case #2 identified carbonate minerals, such as vaterite and hydromagnesite, which were not present before the experiment.