Abstract
Background: Gas exsolution from supersaturated water injection (SWI)
into porous media is of primary interest and has important applications
(CCS, residual NAPL-remediation, CO2-enhanced oil recovery). Bubble
formation and growth kinetics are typically studied in 2D-micro-models
(sandstones-analoges; Zuo et al., 2013; Zuo and Benson, 2013). The only
3D-study was conducted by Li et al., 2017, using the NMR-method, and
indicated a strong dependency of the exsolution process on the initial
CO2-gas phase. However, the NMR-method was not able to attribute these
different fluid-fluid-saturations to key parameters of the dissolution
process such as gas cluster morphology and gas cluster size
distribution. Methods: We conduct a series of column experiments to
study the gas exsolution and gas cluster formation in 3D-porous media
(natural sands, glass beads) using μ-CT. Based on this high-resolution
non-invasive visualization method followed by image processing, we
quantify (i) gas-cluster morphology, (ii) gas-cluster size distribution,
(iii) correlation between pore structure and bubble formation, and (iv)
the impact of surface roughness on exsolution efficiency. Results: We
found that CO2-saturated water equilibrated under ambient pressure, pCO2
= 1.013 bar, (no supersaturation was measured and no pressure reduction
was applied), already leads to gas exsolution of significant amount
(about 10–12% gas saturation) in the presence of untreated
SiO2-surfaces (natural sands, glass beads) which exhibit heterogeneous
wettability. To the best of our knowledge this exsolution phenomenon was
not observed before, and we assume it to be caused by fluid-rock
interactions, i.e. by hydrophobic nucleation sites at the siliceous
surface. The heterogeneous wettability has a dramatic impact on
capillary trapping efficiency and was experimentally observed in
Glass-beads monolayer (Geistlinger and Ataei, 2015, Influence of the
heterogeneous wettability on capillary trapping in glass-beads
monolayers: Comparison between experiments and the invasion percolation
theory, J. Colloid Interface Sci., 459, 230 - 240).