Hydraulic fracturing of certain low permeability, gas shales is performed by injecting tens of thousands of barrels of water and several hundred thousand pounds of proppant at pressures in excess of the minimum total principal stress. A complex fracture network is created. This network is controlled by inherent heterogeneity - including layer interfaces and natural fractures - as has been evidenced by microseismic measurements. The resultant fracture system is far too complicated to forecast in detail, but it can be visualized/approximated as a primary bi-winged planar fracture and a complex secondary fracture network emanating from these two wings. Where injected proppant goes and exactly what the proppant does can only be speculated on at this time. Proppant saltation and duning mechanisms have been recognized for decades and have been demonstrated in physical models. The concept of overflushing (displacing proppant into the fracture with water alone at the end of a fracturing treatment) serves as a platform for reviewing the role of proppant in these low permeability complex systems. This paper reviews such phenomena and standard philosophy, generally considering two-winged, planar fractures and speculates based on experience and physical laws as to where the proppant goes and what proppant "may or may not do" during stimulation of tight gas shales. This paper is a literature review and a thought-building exercise. This combination of a literature review and speculation highlights some of the current uncertainties in the role of proppant for stimulating tight shale formations.
1. INTRODUCTION
The motivation for this technical note was a planning session to design a hydraulic fracture stimulation program. A colleague explained one of the latest fads in fracture designs - overflushing. Proppant is pumped with fracturing fluids so that after injection has stopped there will be physical support to maintain a conductive channel or network to the wellbore. After the designed amount of proppant has been pumped, a final flush stage is pumped to displace any proppant from the wellbore yet not remove near-wellbore support near the well. Overflushing means that the volume of this final stage of unpropped fluids exceeds the volume of the tubulars in the well that the injection is pumped down. Historically, overflushing has been regarded as a bad idea (refer to Figure 1). This paper started as an evaluation to determine if overflushing is always inappropriate. It expanded to look at the literature that characterizes proppant transport during hydraulic fracturing and to speculate on how proppant functions in fracturing low permeability gas-bearing shales. Woodworth and Miskimins, 2007 [1], presented a clear summary of how proppant is transported laterally and settles vertically during fracturing. They described some of the earlier physical modelling that has helped clarify mechanisms of proppant movement in fractures. "Kern et al. [2] injected sand and water through two Plexiglas plates at constant rate. They observed the sand transport at different rates. ... Proppant falls to the bottom and begins to build a mound of proppant. The proppant mound develops and rises until the gap between the top of the mound and the top of the fracture reaches a critical value.
