Probing multiscale dissolution dynamics in natural rocks through microfluidics and compositional analysis.
Bowen LingMo SodwatanaArjun KohliCynthia M RossAdam JewAnthony R KovscekIlenia BattiatoPublished in: Proceedings of the National Academy of Sciences of the United States of America (2022)
Mineral dissolution significantly impacts many geological systems. Carbon released by diagenesis, carbon sequestration, and acid injection are examples where geochemical reactions, fluid flow, and solute transport are strongly coupled. The complexity in these systems involves interplay between various mechanisms that operate at timescales ranging from microseconds to years. Current experimental techniques characterize dissolution processes using static images that are acquired with long measurement times and/or low spatial resolution. These limitations prevent direct observation of how dissolution reactions progress within an intact rock with spatially heterogeneous mineralogy and morphology. We utilize microfluidic cells embedded with thin rock samples to visualize dissolution with significant temporal resolution (100 ms) in a large observation window (3 × 3 mm). We injected acidic fluid into eight shale samples ranging from 8 to 86 wt % carbonate. The pre- and postreaction microstructures are characterized at the scale of pores (0.1 to 1 µm) and fractures (1 to 1,000 µm). We observe that nonreactive particle exposure, fracture morphology, and loss of rock strength are strongly dependent on both the relative volume of reactive grains and their distribution. Time-resolved images of the rock unveil the spatiotemporal dynamics of dissolution, including two-phase flow effects in real time and illustrate the changes in the fracture interface across the range of compositions. Moreover, the dynamical data provide an approach for characterizing reactivity parameters of natural heterogeneous samples when porous media effects are not negligible. The platform and workflow provide real-time characterization of geochemical reactions and inform various subsurface engineering processes.
Keyphrases
- deep learning
- heavy metals
- high throughput
- multiple sclerosis
- induced apoptosis
- mass spectrometry
- electronic health record
- ms ms
- single cell
- signaling pathway
- oxidative stress
- machine learning
- artificial intelligence
- ultrasound guided
- circulating tumor cells
- drinking water
- molecular dynamics simulations
- density functional theory
- label free