Quantifying Non-API Proppant Transport and Fracture Permeability as a Function of Solids Loading and Particle Shape

The use of non-API proppant in hydraulic fracturing has the potential to significantly reduce the cost of hydraulic fracturing operations. However, the influence of particle angularity on proppant transport in rough fractures is not well understood. This work utilises direct numerical simulation (DNS) to quantify changes in fracture permeability as a function of proppant aspect ratio, solid volume fraction, and fracture width. High-fidelity simulations of proppant transport through realistically-textured fractures were conducted using an established numerical framework based on the lattice Boltzmann method and discrete element methods. Particle angularity was captured via clumps of overlapping spheres and the fracture surfaces were described by a power spectral density. The results show that fracture permeability, which is a proxy for injection pressure, smoothly decreases with increasing proppant concentration for all proppant and fracture characterisations. No distinct screen-out was observed, irrespective of particle shape, which contradicts the prevailing assumption that screen-out instantaneously occurs at some threshold of proppant concentration and fracture width. Interestingly, the change in fracture permeability with particle angularity was found to be non-monotonic (decreasing, then increasing, and then decreasing again). It is hypothesized that increases in permeability are a result of the preferential alignment of particles with the flow streamlines. This requires further investigation, along with the analysis of more complex clump shapes such as triangles and tetrahedra. The developed permeability-concentration curves clearly show the influence of particle angularity on injection pressure and can be used to help inform operational decision making when non-API proppant is proposed in reservoir stimulation. 1 INTRODUCTION The American Petroleum Institute (API) defines criteria for the classification and quality control testing of sand used in hydraulic fracturing. These metrics include sieve analysis for size distribution, sphericity and roundness (minimum > 0.6), acid solubility, turbidity, and crush resistance. The imperative to reduce well completion costs has led to the exploration and deployment of lower-cost products which do not meet one or more of the API criteria. In some limited scenarios, the properties of non-API proppant are deliberately sought (e.g. the use of angular, non-spherical particles to increase pack stability in the fracture due to locking, thereby reducing flow-back to the well). An important consideration associated with the use of non-API proppant is the effect on stimulation efficacy and estimated ultimate recovery (EUR).