Physics-Based Multiscale Modeling of Proppant Transport Across Fracture Intersections in Naturally Fractured Carbonate Reservoirs

Aisha Labak, Suleiman Hassan, Sonny Irawan, Shams Kalam, Randy Doyle Hazlett, Shanker Krishna

2026

Closed access Metadata (CSL-JSON)
DOI 10.2118/233277-ms
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

Abstract This study develops a physics-based multiscale framework to predict proppant transport and placement in naturally fractured carbonate reservoirs by linking grain-scale fracture-intersection behavior to reservoir-scale Discrete Fracture Network (DFN) performance. Computational Fluid Dynamics and Discrete Element Method (CFD-DEM) simulations were conducted for straight fractures, T-junctions, X-junctions, rough T-junctions, and proppant-loading-varied T-junction cases. The extracted descriptors included proppant retention factor, branch split ratio, pressure-drop response, conductivity modifier, and aperture-to-particle bridging behavior. These descriptors were then upscaled into a carbonate DFN to evaluate how local retention, diversion, and conductivity loss affect proppant coverage, effective connectivity, and flow capacity. Numerical verification confirmed mesh and time-step independence, with mass-balance error not exceeding 0.94%. The straight-fracture case showed limited retention (Rp = 0.038), whereas the T-junction, X-junction, and rough T-junction increased retention to 0.269, 0.283, and 0.349, respectively. Branch entry in T-junction cases ranged from 0.308 to 0.328 of the mobile proppant, and the rough T-junction produced the maximum pressure-drop increase of 59.1%. The conductivity modifier ranged from 0.629 to 0.721 across the intersection and proppant-loading cases, while partial bridging or sustained accumulation was observed at Λ = 2.46. Coupling the CFD-DEM descriptors into the DFN reduced the sufficiently propped fracture fraction from 0.70 to 0.45, normalized flow capacity from 1.00 to 0.73, and effective propped connectivity from 0.82 to 0.56. Sensitivity analysis identified roughness, aperture variability, and intersection density as the strongest controls on proppant placement. The framework shows that geometric connectivity alone can overestimate effective proppant placement and that intersection-scale transport descriptors improve the evaluation of network-scale stimulation.