11:15 AM - 11:30 AM
[HSC06-09] DEM simulation of multidirectional hydraulic fracturing achieved by shear thickening fluid
Keywords:DEM, Fracturing, STF
The conventional hydraulic fracturing method used in subsurface resource development has a fundamental limitation: it can only induce fractures along the maximum principal stress direction, thereby restricting permeability enhancement to a single orientation. To overcome this limitation, shear thickening fluid (STF) has been proposed as a fracturing fluid (Mukuhira et al., ARMA, 2024), showing its potential to generate multi-directional fractures. STF is a type of non-Newtonian fluid and its viscosity varies with the shear rate; it shows quasi-solid behavior under shear. Laboratory experiments in the previous study have demonstrated multiple pressure build-ups and drops, as well as fractures forming in various directions, which are phenomena absent in conventional hydraulic fracturing. However, the underlying mechanism of these behaviors has not been fully understood due to the limitations of experimental observation. To address this, we conducted numerical simulations using the discrete element method (DEM) to model STF fracturing processes.
In this study, we implemented a shear rate-dependent viscosity function into our in-house 2D DEM code which can simulate hydraulic fracturing. Specifically, we implemented a shear rate-dependent viscosity model, allowing STF viscosity to vary dynamically in response to local flow velocities. Additionally, a relaxation time function was introduced to account for the time-dependent recovery of viscosity.
Simulations were performed using a rock model with a single pre-existing fracture to analyze the effects of STF on pressure variation and fracture propagation. The 2D square rock models had a borehole in the center and the pre-existing fracture set from the borehole to the one boundary of the model. In the calculation, STF and Newtonian fluid were injected into the borehole of each model for comparison, and then borehole pressure variation and subsequent phenomena were observed.
The simulation results demonstrated that STF injection led to higher borehole pressure than the case of Newtonian fluid and an additional fracture generation with a pressure drop. In the case of Newtonian fluid, the borehole pressure linearly increased and then stabilized at a constant level resulting in a steady-state flow condition. Since the fluid just flowed out of the pre-existing fracture, additional fracture generations did not occur in the case of Newtonian fluid. The analysis of the results showed a drastic increase in viscosity around the borehole in the case of STF.
We interpret the results as follows. The increase in viscosity near the borehole reduced the fluid outflow from the borehole into the pre-existing fracture and led to an increase in borehole pressure to a higher value than that of the Newtonian fluid case. Higher borehole pressure may cause additional fracture generation which did not occur in the case of Newtonian fluid.
The simulation results demonstrated our successful implementation of the STF shear thickening effect in our DEM, as well as the qualitative reproduction of STF's quasi-solid behavior. In addition, the results provide insights into the mechanism of multidirectional fracturing by STF, that high viscous STF acted as a seal within fractures, temporarily restricting flow and promoting borehole pressure build-up. Our simulations support the hypothesis in the previous study (Mukuhira et al., ARMA, 2024) that STF viscosity variations play a crucial role in multi-directional fracture propagation.
This study provides new insights into the mechanism of STF fracturing and demonstrates the effectiveness of our modified DEM model. By incorporating a shear rate-dependent viscosity model and relaxation time function, we were able to capture the dynamic behavior of STF and its effects on fracture propagation. Future work will focus on refining the model to reproduce the whole process of STF fracturing and fully understand its mechanism.
In this study, we implemented a shear rate-dependent viscosity function into our in-house 2D DEM code which can simulate hydraulic fracturing. Specifically, we implemented a shear rate-dependent viscosity model, allowing STF viscosity to vary dynamically in response to local flow velocities. Additionally, a relaxation time function was introduced to account for the time-dependent recovery of viscosity.
Simulations were performed using a rock model with a single pre-existing fracture to analyze the effects of STF on pressure variation and fracture propagation. The 2D square rock models had a borehole in the center and the pre-existing fracture set from the borehole to the one boundary of the model. In the calculation, STF and Newtonian fluid were injected into the borehole of each model for comparison, and then borehole pressure variation and subsequent phenomena were observed.
The simulation results demonstrated that STF injection led to higher borehole pressure than the case of Newtonian fluid and an additional fracture generation with a pressure drop. In the case of Newtonian fluid, the borehole pressure linearly increased and then stabilized at a constant level resulting in a steady-state flow condition. Since the fluid just flowed out of the pre-existing fracture, additional fracture generations did not occur in the case of Newtonian fluid. The analysis of the results showed a drastic increase in viscosity around the borehole in the case of STF.
We interpret the results as follows. The increase in viscosity near the borehole reduced the fluid outflow from the borehole into the pre-existing fracture and led to an increase in borehole pressure to a higher value than that of the Newtonian fluid case. Higher borehole pressure may cause additional fracture generation which did not occur in the case of Newtonian fluid.
The simulation results demonstrated our successful implementation of the STF shear thickening effect in our DEM, as well as the qualitative reproduction of STF's quasi-solid behavior. In addition, the results provide insights into the mechanism of multidirectional fracturing by STF, that high viscous STF acted as a seal within fractures, temporarily restricting flow and promoting borehole pressure build-up. Our simulations support the hypothesis in the previous study (Mukuhira et al., ARMA, 2024) that STF viscosity variations play a crucial role in multi-directional fracture propagation.
This study provides new insights into the mechanism of STF fracturing and demonstrates the effectiveness of our modified DEM model. By incorporating a shear rate-dependent viscosity model and relaxation time function, we were able to capture the dynamic behavior of STF and its effects on fracture propagation. Future work will focus on refining the model to reproduce the whole process of STF fracturing and fully understand its mechanism.