No abstract

Fracture fluid flow back has been identified as one of the major challenges of hydraulic fracturing operations conducted in shale reservoirs. Factors causing the very low fracture fluid recovery need to be well understood and properly addressed, in order to get full benefits from costly hydraulic fracture jobs conducted in unconventional reservoirs. Despite the recent surge of investigations of the problem, one major question still remains: what happens to the fracture fluid that is not recovered? Does it stay in the fracture or does it go into the matrix? In case of both mechanisms are responsible for fracture fluid retainment, what fraction of fracture fluid stays in the propped fracture and what fraction is transferred from fracture to matrix. The focus of the current study is to understand if the transfer of fracture fluid from fracture to matrix through imbibition is of significant importance. We systematically measure the imbibition rate of water, brine, and oil into the actual core samples from the three shale sections of Horn River basin (i.e., Fort Simpson, Muskwa and Otter Park). We characterize the shale samples by measuring, porosity, wettability, mineral composition through XRD analysis, and interpreting the well log data. The results show that imbibition could be a viable mechanism for fluid transfer from fracture to matrix in Horn River shales. The comparative study shows the imbibition rate in the direction parallel to the bedding plane is higher than that in the direction perpendicular to the bedding. The study also suggests that the imbibition rate of the aqueous phases is significantly higher than that of the oleic phases.

The imbibition of fracturing fluid into the shale matrix is identified as one of the possible mechanisms leading to high volumes of water loss to the formation in hydraulically fractured shale reservoirs. In an earlier study (Makhanov et al., 2012), several spontaneous imbibition experiments were conducted using actual shale core samples collected from Fort Simpson, Muskwa and Otter Park formations, all belong to the Horn River shale basin. This study provides additional experimental data on how imbibition rate depends on type and concentration of salts, surfactants, and viscosifiers. The study also proposes and applies a simple methodology to scale up the lab data for field-scale predictions. The data shows that an anionic surfactant reduces the imbibition rate due to the surface tension reduction. The imbibition rate is even further reduced when KCl salt is added to the surfactant solution. Surprisingly, viscous XG solutions show a considerable spontaneous imbibition rate when exposed to organic shales, although their viscosity is much higher than water viscosity. This observation indicates that water uptake of clay-rich organich shales is mainly controlled through preferential adsorption of water molecules by the clay particles, and high bulk viscosity of the polymer solution can only partly reduce the rate of water uptake. The field scale calculations show that water loss due to the spontaneous imbibition during the shut-in period is a strong function of fluid/shale properties, fracture-matrix interface, and soaking time. The presented data and analyses can be used to explain why some fractured horizontal wells completed in gas shales show an immediate gas production after extended shut-in periods.