Criterion of proppant pack mobilization by filtrating fluids: Theory and experiments

The goal of hydraulic fracturing is placing a granular material, proppant, in the created fracture to sustain its conductivity and attain economical production rates. After fracture closure on the placed proppant and subsequent pressure drawdown in a well, the consolidated proppant can be dragged by the producing fluids and undesirably be produced back to the well. Rough estimations or accurate simulations of this phenomenon require a valid quantitative criterion of proppant mobilization in a fracture. In this work, we analytically derive such criterion from the fundamental principles. The criterion is expressed in terms of the minimum fluid filtration velocity that destabilizes the proppant. It is sensitive to such important factors as cohesion, thickness and permeability of a proppant pack, mean grain size, roughness and softness of fracture walls, and filtrating fluid viscosity. The developed theory considers the existing unloaded edge of a proppant pack, its granular structure, and related finite cohesion of particles holding each other in a pack. These key properties are shown to play a dominant role for mobilization of a proppant in hydraulic fractures. On the other hand, we conduct an extensive series of laboratory experiments mimicking subterranean fracture closure on proppant, fluid filtration through it, and registering proppant flowback. In the experiments, we try a wide range of commercial proppants, less- and more-viscous fluids, and stiff and soft fracture walls. We associated proppant flowback onset with the change of proppant pack height. We also show how proppant bridging depends on the particle embedment into soft walls and illustrate various geometries of proppant washing out, giving clues to various scenarios of proppant flowback evolution. Finally, we observe a good correspondence of our theory and experimental observations. Theoretical hypotheses of unstressed edge and finite cohesion of a granular solids serve as the only means to explain the observed range of filtration velocity magnitudes for proppant mobilization and their independence from the value of confining stress. The reciprocity between the cohesion and critical velocity revealed by the theory is confirmed by experiments. It allows measuring only one of them to estimate another. •Critical filtration velocity for proppant pack mobilization is derived and validated.•Mobilization of a compressed pack occurs at the edge and independent of the stress.•Pack stability depends on