Abstract

To gain understanding of the mechanisms and rock properties which dominate fracture conductivity generation and decline, fracture conductivity behavior was investigated in several shale reservoirs in the laboratory to identify the main parameters that control the fracture conductivity. The comprehensive study investigated the conductivity behaviors of four shale plays, the Barnett, Fayetteville, Marcellus and Eagle Ford, by testing fracture conductivity as a function of closure stress using collected outcrop rock samples. Conductivity of fractured rock samples are measured using a modified API RP 61 procedure. In addition to conductivity, a plethora of rock properties such as Young's modulus, Poisson's ratio, Brinell hardness, rock brittleness, and mineralogy are measured corresponding to each conductivity test sample. Other parameters considered in this study include proppant mesh size, proppant concentration, and fracture topography to see the resulting impact on conductivity. With all measurements, a multivariate analysis is carried out with the goal of seeking a correlation between fracture conductivity and the influential rock properties and treatment parameters. All results are cross-compared to identify similarities and differences between formations as well as the most important parameters influencing fracture conductivity relative to each formation.

The results indicate a positive trend between both Brinell hardness and Young's Modulus with fracture conductivity. Increased fracture face surface roughness is also found to have an overall decrease in conductivity in propped fractures. The influential parameters are different for different shale plays. Some parameters are more important in one formation but less influential in others. The observations are integrated into a preliminary simple correlation for conductivity behavior. The observations and results from this study clarify the similarity and differences in varying shale plays.

You can access this article if you purchase or spend a download.