CFD-DEM modeling of fracture initiation with polymer injection in granular media

TL;DR

This work addresses fracture initiation in granular media caused by injection of shear-thinning polymer solutions. It employs a CFD-DEM coupling with a power-law rheology to capture fluid–particle interactions and validates the model against laboratory data, demonstrating consistent trends in peak pressure, confining stress, and dimensionless fracture indicators. Key findings show that fluid rheology, injection rate, and solid modulus jointly control fracture morphology—Guar induces grain-displacement fractures, XG promotes infiltration-dominated fractures, and stiffer media produce narrower fracture zones—while a threshold-based criterion using $1/Π_1$ and $τ_2$ predicts initiation. The proposed criteria offer a practical tool for predicting fracture initiation under non-Newtonian polymer injection and highlight the role of micromechanical properties in fracture outcomes, with implications for designing fracturing and remediation strategies.

Abstract

We numerically study mechanisms and conditions of fracture initiation in granular material induced by non-Newtonian polymer solutions. A coupling approach of computational fluid dynamics and discrete element method is utilized to model the fluid flow in a porous medium. The flow behavior of polymer solutions and drag force acting on particles are calculated based on a power-law model. The adequacy of the numerical model is confirmed by comparing the results with a laboratory experiment. The numerical results are consistent with the experimental data presenting similar tendencies in dimensionless parameters that incorporate fluid flow rate, rheology, peak pressure, and confining stress. Results show that fluid flow rate, rheology, and solid material characteristics strongly influence fracture initiation behavior. Injecting a more viscous guar-based solution results in wider fractures induced by a grain displacement. A less viscous XG-based solution creates more linear fractures dominated by an infiltration. The peak pressure ratio between two fluids is higher in rigid material compared to softer material. Finally, the dimensionless parameters $1/Π_1$ and $τ_2$, which consider fluid and solid material properties accordingly, are good indicators in determining fracture initiation induced by shear-thinning fluids. Our numerical results show that fracture initiation occurs above $1/Π_1 = 0.06$ and $τ_2 = 2\cdot 10^{-7}$.

Figures (9)

• Figure 1: Numerical setup of the validation and fracture initiation models.
• Figure 2: Permeability calculation simulation setup.
• Figure 3: (a) Time history of injection pressure (b) $P_{peak}/\sigma_3$ versus confining stress $\sigma_3$.
• Figure 4: Comparison of dimensionless parameters between the experimental data and numerical results of guar-based polymer solution at $U = 5\cdot10^{-4}$ m/s and $\sigma_3$ = 500, 750, 1000, 1250 and 1500 Pa.
• Figure 5: Comparison of fracture initiation by injecting different fluids.