IONIZATION IN “WET” CO2­ DETERMINED BY CONDUCTIVITY MEASUREMENTS FROM 298 TO 473K AND 8 TO 20MPA

Geoengineering projects such as CO₂ sequestration and enhanced geothermal energy production involve injection of CO₂ into deep formations. Injection of CO₂ into these formations will result in the dissolution of some water into the CO₂. Recent experimental studies suggest that water-bearing (“wet”) carbon dioxide is highly reactive, and may react with minerals and other materials, but the mechanism(s) behind these reactions are not well understood. If “wet” CO₂ is as reactive recent studies show, it could corrode equipment such as drill pipes and turbines in power plants and impact CO₂storage security.

Several mechanisms have been proposed to explain the reactivity of “wet” CO₂. Two of these mechanisms involve ionization of carbonic acid, either in the bulk fluid or within a thin film at the fluid-solid interface. Other possibilities include direct interaction of molecular CO₂ and/or H₂O with certain materials. In this study we determined the degree to which ionization in the bulk fluid might explain the reactivity of “wet” CO₂using a flow-through conductivity cell. Our conductivity measurements are sensitive to ionization in a bulk fluid but insensitive to thin films and neutral species.

We found no relationship between conductivity and the concentration of H₂O in CO₂ (from <1 ppmw to saturation) at 298 K and 8 MPa. The conductivity of “wet” CO₂ was indistinguishable from “dry” CO₂, and the total conductivity was much less than that of ultra-pure water at ambient conditions. From 298 to 473 K and 8 to 20 MPa, there was no detectible difference between “dry” and “wet” CO₂ (up to 1500-1600 ppmw H₂O). The total conductivity of these solutions in some cases is comparable to the apparent conductivity of vacuum.

The absence of measureable conductivity in “wet” CO₂ indicates that the concentration of mobile ions is vanishingly small at these PTX conditions, which include conditions where “wet” CO₂ has been shown to be highly reactive. We confirm the predictions of prior workers that ionization of carbonic acid in bulk “wet” CO₂ (if any) is not sufficient to explain observed rates of corrosion at these PTX conditions.