Abstract:

Fluid transport during hydraulic fracturing can be either proppant or asperity dominated. In the absence of proppants, dilatancy of the natural fractures during shear reactivation is required to provide sufficient aperture and hydraulic conductivity. If a hydraulic fracture intersects a natural fracture, the fracturing fluid enters the natural fractures, increasing pore pressure and decreasing the effective normal stress on the fracture plane. If the effective normal stress becomes low enough, the natural fracture may slip and dilate, resulting in increased aperture and hydraulic conductivity. To understand these processes, we conducted direct shear tests on rock fractures with constant normal load and examined the effect of slip and dilatancy of asperities on mechanical aperture. Using 3D coordinates of the surface asperities and measuring shear displacement and dilation during shear testing, the evolution of the mechanical aperture was calculated as a function of slip and normal stress. The calculation suggested that when effective normal stress is low, small slip events along a rough surface can induce dilatancy along the fracture surface that will cause considerable increases in the hydraulic aperture of the fracture.

Introduction

Proppants are introduced during hydraulic fracturing to increase fracture transmissivity during well production (Cipolla et al. 2009). But can fracture conductivity also be increased without proppant, for example, by water fracing? (Mayerhofer et al 1997). In contrast to conventional propped hydraulic fracture treatments, water fracs rely on reactivation of natural fractures to induce permanent shear induced dilation, which enhances reservoir permeability (Chen et al 2000, Weng et al 2015). Although the conductivity of un-propped shear-induced fractures is relatively low compared to that of the propped fractures, such conductivity can still play an important role in enhancing the productivity of ultra-low-permeability rocks like shale (Zhang et al. 2013, Weng et al 2015, Jansen et al. 2015).

This content is only available via PDF.
You can access this article if you purchase or spend a download.