Integrated Physics-Based Modeling and Microfluidics for Quantifying Multiphase Carbonate Dissolution in Rocks

Acid dissolution of carbonate formations is critical to the energy transition and relevant to many engineering applications. The dynamics of the dissolution reaction are complex, strongly depend both on the flow properties and sample mineralogy and are further complicated by the production of carbon dioxide gas bubbles from the reactive surface, which renders the system multiphase. Quantifying the impact of multiphase flow conditions on effective reaction rates of carbonate dissolution has challenged experimental methods focused on core-based characterization techniques. In this work, we use microfluidic devices that contain carbonate-rich (86 wt%) rock samples with a cylindrical shape to observe their dissolution upon injection of hydrochloric (HCl) acid under both single and multiphase conditions. The dissolution reaction is visualized and recorded at high temporal resolution using a high-speed camera and is quantified through machine learning (ML)-based image segmentation. First, we combine ML-enabled image analysis with physics-based modeling to estimate the instantaneous reaction rates of carbonate dissolution under single-phase conditions and validate that it follows a first-order reaction rate law. Then, we use the proposed approach to determine the effective dissolution rate under multiphase flow conditions, i.e. when - at higher HCl concentration - the formation of CO$_2$ bubbles shields the adjacent carbonate surface hindering reaction progress. We find that, under such conditions, the effective reaction rate decreases by one order of magnitude, strongly deviating from the reaction rate law previously determined for single-phase conditions and that current models are not able to capture the impact of gas shielding effects on effective reaction rates under multiphase flow conditions. We also find that the natural chemical heterogeneity of rocks leads to the in situ formation of an unreacted mineral porous layer which serves as the substrate for gas bubbles to nucleate and grow, which changes the conceptual model established for calcite systems.