Answers are grounded in this paper and cite their sources.
Examines CO2 viscous fingering during immiscible flooding, suggesting improved front characterization methods.
This study investigates viscous fingering during CO2 immiscible flooding under diffusion-dissolution effects by combining two-dimensional (2D) visualization experiments with an analytical front-characterization model. A 25 cm x 25 cm visual flat-plate model packed with 60-mesh quartz glass beads (250 micrometre diameter) was used, and CO2 was injected at 0.25 mL/min under 28 °C and 2 MPa back pressure. The observed fingering process can be divided into three quantitative stages: an initial scattered-channel stage at 0.05–0.15 PV, a dendritic-growth stage at 0.15–0.30 PV, and a connected network stage at 0.30–0.40 PV. Image-based evaluation shows that the CO2 swept area increased nonlinearly with injected pore volume and reached a final sweep efficiency of 32.62% at 0.40 PV, while high-oil-saturation zones remained between preferential gas channels. The proposed model couples Fickian CO2 diffusion, concentration-dependent oil viscosity, mobility-ratio variation, and a dimensionless fingering variation rate, thereby linking front instability to diffusion-dissolution-induced viscosity redistribution. The model predicts a monotonic increase in fingering variation rate along the displacement direction (σ L > σ 0), which is consistent with the experimentally observed transition from localized penetration to a dendritic fingering network. Compared with previous visual or numerical descriptions of immiscible fingering, the novelty of this work lies in introducing a front-oriented characterization index for CO2 immiscible flooding that explicitly incorporates diffusion-dissolution effects and can be connected with image-derived sweep efficiency. The model is intended as a first-order, front-oriented descriptor for homogeneous, horizontal, viscosity-dominated systems and should be calibrated with relative-permeability and PVT data before field-scale use.
No takes yet. Share an insight, caveat, or question.
Peng et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: