Research

It is well known that fluid injection (e.g., CO₂ sequestration, waterflooding, or subsurface stimulation) can induce seismic events. The mechanisms for injection-time seismicity are relatively well established, typically attributed to either effective stress reduction due to elevated pore pressure or increased total stress that reactivates pre-existing fractures. However, a significant portion of induced seismic events, particularly those with larger magnitudes, occur after injection has ceased. The mechanisms driving these post-injection events remain poorly understood. Our laboratory is actively investigating this knowledge gap using advanced numerical modeling tools.

Hydraulic fracturing has revolutionized the oil and gas industry, yet its quasi-static nature constrains fracture propagation to the direction of minimum in situ stress. For applications such as in-situ mining, geothermal energy extraction, geological hydrogen stimulation, and emerging technologies like pulsed-power fracturing, alternative approaches are needed to mitigate short-circuiting, enhance permeability, and increase geochemical reaction surfaces by orders of magnitude. The Geosystem Innovation Laboratory (GIL) is addressing these challenges by developing advanced numerical tools to model the coupled poro-dynamic fracturing process.

Hydraulic fracturing has revolutionized the oil and gas industry, and decades of field practice have enabled a deep understanding of many aspects of the process. However, several key phenomena remain insufficiently addressed:

1. The conventional mode-I fracture propagation mechanism does not explain the observed generation of fracture swarms;
2. Modeling of fracture growth coupled with proppant transport and settlement requires significant improvement;
3. Fracture closure during shut-in and flowback periods—and the resulting residual fracture aperture—are still not well characterized.

Beyond the inherent complexity of these multiphysics processes, achieving high-fidelity modeling of hydraulic fracturing demands extensive computational resources. Our group is developing advanced computational tools, incorporating refined physical mechanisms and GPU-accelerated simulations, to tackle these challenges.

Geothermal battery energy storage, using geological formations to store energy in the form of thermally heated brine, has been proposed as a solution to balance the increasing and intermittent nature of renewable energy generation and to enhance the stability of U.S. power grids. This concept has already been successfully applied for building heating and data center cooling by storing excess heat or cold during periods of low demand and recovering it during peak demand. This research aims to:

• Identify suitable subsurface formations for geothermal energy storage,
• Quantify storage efficiency within an integrated energy storage framework,
• Develop strategies to mitigate risks associated with subsurface system failures.

Granular material can switch its behavior from solid-like (able to support quasi-static shear loads) to liquid-like (it can flow in a dense state). This research aims to

• Develop and validate physics-based constitutive laws for biomass granular flow behavior;
• Model biomass granular flow and utilize simulation results to control unit operation conditions and optimize equipment design.

Brittle materials—such as concrete, rock, and ceramic composites—exhibit complex mechanical behavior at the meso-scale, including:

• Stress-induced damage and stiffness anisotropy,
• Nonlinear stress–strain relationships,
• Volumetric dilation caused by irreversible strain,
• Unilateral effects due to crack closure,
• A brittle-to-ductile transition under increasing confining stress,
• Distinct mechanical responses in tension and compression.

Physically, these effects arise from the nucleation and propagation of microcracks at grain boundaries and cross pore spaces. The growth and coalescence of these diffuse microcracks lead to the formation of localized macro-fractures and ultimately to structural failure. Accurately modeling this cross-scale process has been a long-standing challenge. The goal of this project is to:

• Formulate constitutive damage models that captures these behaviors using physically meaningful and identifiable material parameters,
• Develop algorithms and a theoretical framework to model the transition from microcrack evolution to macro-fracture formation,
• Implement the constitutive and cohesive fracture models within finite element codes using appropriate discretization techniques,
• Apply the computational framework to perform complete failure analyses of materials such as concrete, granite, and shale.