CO2–Water–Rock Wettability: Variability, Influencing Factors, and Implications for CO2 Geostorage

Abstract

Carbon geosequestration (CGS) has been identified as a key technology to reduce anthropogenic greenhouse gas emissions and thus significantly mitigate climate change. In CGS, CO2 is captured from large point-source emitters (e.g., coal fired power stations), purified, and injected deep underground into geological formations for disposal. However, the CO2 has a lower density than the resident formation brine and thus migrates upward due to buoyancy forces. To prevent the CO2 from leaking back to the surface, four trapping mechanisms are used: (1) structural trapping (where a tight caprock acts as a seal barrier through which the CO2 cannot percolate), (2) residual trapping (where the CO2 plume is split into many micrometer-sized bubbles, which are immobilized by capillary forces in the pore network of the rock), (3) dissolution trapping (where CO2 dissolves in the formation brine and sinks deep into the reservoir due to a slight increase in brine density), and (4) mineral trapping (where the CO2 introduced into the subsurface chemically reacts with the formation brine or reservoir rock or both to form solid precipitates). The efficiency of these trapping mechanisms and the movement of CO2 through the rock are strongly influenced by the CO2–brine–rock wettability (mainly due to the small capillary-like pores in the rock which form a complex network), and it is thus of key importance to rigorously understand CO2-wettability. In this context, a substantial number of experiments have been conducted from which several conclusions can be drawn: of prime importance is the rock surface chemistry, and hydrophilic surfaces are water-wet while hydrophobic surfaces are CO2-wet. Note that CO2-wet surfaces dramatically reduce CO2 storage capacities. Furthermore, increasing pressure, salinity, or dissolved ion valency increases CO2-wettability, while the effect of temperature is not well understood. Indeed theoretical understanding of CO2-wettability and the ability to quantitatively predict it are currently limited although recent advances have been made. Moreover, data for real storage rock and real injection gas (which contains impurities) is scarce and it is an open question how realistic subsurface conditions can be reproduced in laboratory experiments. In conclusion, however, it is clear that in principal CO2-wettability can vary drastically from completely water-wet to almost completely CO2-wet, and this possible variation introduces a large uncertainty into trapping capacity and containment security predictions.