Unraveling the Scale Effect on Fracture Plugging: A Quantitative Study from Lab-Scale Slots to Field-Scale Fractures via Hydrodynamic Analysis

Abstract Conventional core-scale laboratory fracture-plugging tests may overestimate the ability of granular lost circulation materials (LCMs) to form a complete plugging zone under field-scale conditions. This discrepancy arises because the effects of increasing fracture length and height on long-distance particle transport, plugging behavior, and plugging-zone evolution remain insufficiently quantified. Accordingly, a multiscale visualization apparatus was developed with micro- (5 × 1 cm), meso- (20 × 10 cm), and macro-scale (100 × 30 cm) fracture modules. Each module had a tapered aperture decreasing from 3 mm at the inlet to 1 mm at the outlet. The experiments used a 0.5 wt % carboxymethyl cellulose (CMC) suspension, nylon particles of different sizes, and calcium carbonate particles. High-speed imaging, pressure monitoring, and dimensionless analysis were combined to investigate the effects of fracture scale on particle transport, bridging, settling, and plugging. At corresponding cross sections, the generalized fluid Reynolds numbers (Re) differed by no more than 17.4% among the three scales, and all flows remained within the laminar regime. Nevertheless, for a given particle size, the dimensionless bridging initiation time (DBIT) increased by 8.8–13.3 times from the micro- to the macro-scale fracture. For the 0.9 mm nylon particles, plugging coverage (PC) decreased from 98.0% to 14.6%, whereas local packing density (PD) remained within 80–90%. Scale enlargement, therefore, primarily weakened the spatial connectivity of the particle skeleton rather than the compactness of local deposits. The settling time scale ratios (STR) of the calcium carbonate particles were 0.577, 0.231, and 0.384 in the micro-, meso-, and macro-scale fractures, respectively. At the microscale, the bottom deposit expanded across the full fracture height. At the mesoscale, deep multipoint bridging developed into a fracture-spanning skeleton. At the macroscale, a mobile, dune-like bed coexisted with an open upper bypass channel. A semiempirical scaling correction framework was further developed using the flow-path length-to-hydraulic-diameter ratio, settling deviation, and coverage connectivity. Comparisons with independent visualization experiments and the computational fluid dynamics-discrete element method (CFD-DEM) results supported its ability to identify consistent trends. In an application to Well Z229, a field pressure-bearing capacity of 7.76 MPa was achieved. These results indicate that, under comparable laminar-flow conditions, scale effects on fracture plugging are controlled mainly by transport distance, absolute fracture height, and residence-settling competition. The findings provide guidance for selecting fracture evaluation models, designing particle-size gradations for LCMs, and specifying field squeeze-injection parameters.