Abstract
Laboratory and field studies have demonstrated a strong correlation between the volume of proppant deployed in hydraulic fracturing operations and subsequent reservoir productivity. In recent years, the desire to improve proppant performance has led to the development of new generations of exotic proppants, as well as new propping strategies. Nevertheless, the factors controlling performance of even traditional proppants in real rock fractures are poorly understood. Improved models are needed to help devise optimal strategies for deploying traditional and new varieties of proppant.
Large-scale proppant models frequently rely on empirical closure relationships to represent real-world transport behavior. However, care must be taken when applying such relationships outside their derived context. In particular, most models employ closure relationships determined for slurries where the fluid dimensions vastly exceed the particle size. This is not true for fracture flow, where wall effects alter the effective transport properties and introduce new interaction forces.
This paper describes an ongoing study employing a combination of high fidelity numerical simulations and fracture-scale experiments to describe the transport properties of proppant particles in fractures. The experimental work examines proppant movement in clearplastic three-dimensional reproductions of the shale surfaces recreated using three dimensional printing. These specially tailored flow cells are used in combination with micro-capsules for improved particle tracking in dense particle packs. The same shale surfaces are also employed in high-resolution particulate flow simulations in which both the particles and interstitial fluid motion are explicitly represented. The data gathered from these experiments and simulations are used to help constrain models of settling and dispersion in particle-laden fluids within fractures.
1. INTRODUCTION
Correct proppant placement is key to maintaining fracture permeability following hydraulic stimulation of geological resources, in particular, stimulation of unconventional oil and gas reservoirs. Laboratory studies have demonstrated improvements of several orders of magnitude in the permeability of propped versus unpropped fractures [1], while analysis of production rates in the field show a strong correlation between the amount of proppant introduced and production rates observed in the field [2, 3]. Nevertheless, the nature of proppant transport in natural fractures is poorly understood. Completed wells, for example, can experience flowback of more than 10% of the injected proppant and proppant concentration schedules are often determined through trial and error, rather than design [4, 5, 6]. The desire to improve proppant performance has lead to the proliferation of a wide variety of propping strategies. These include the development of lower density proppants to reduce settling rates and improve transport in low viscosity fluids, and the development of strategies employing multiple modal proppant size distributions to improve propping at the fracture tip. Elsewhere cyclic pumping and variable proppant concentrations are employed in an effort to create heterogeneous proppant deposition [7]. These strategies are not without risk: mixing of proppants of different size can result in a reduction in permeability; low density proppants are more susceptible to flow back and are more costly than traditional proppants; and uncertainties surrounding dispersion characteristics hinders heterogeneous proppant deployment. Without accurate models of proppant transport the correct deployment and efficacy of such strategies is a matter of conjecture
