In-situ imaging of fracture development during maturation of an organic-rich shale: Effects of heating rate and confinement

Abstract

The statistics and dynamics of fractures formed during the accelerated maturation of kerogen-rich shale was investigated by heating Green River Shale (R-8 unit, Piceance Basin, northwestern Colorado) core samples while 3D X-ray microtomographic images were acquired. Previous studies have shown that, when there was no confining stress, fractures formed while the kerogen contained in the shale matured, and the produced hydrocarbon was expelled through these fractures. In the present study, X-ray tomographic scans at multiple voxel sizes were conducted on similar samples during heating. In one experiment, the shale sample was tightly fitted in a non-porous ceramic tube to confine it while it was heated. Three unconfined samples were heated and then held at different final temperatures to investigate the effects of the gas production rate on fracturing. 3D image processing was used to survey fracture network development, and time-lapse 2D digital image correlation analysis was used to monitor the development of the displacement and strain fields. The results revealed that fracturing is strongly dependent on the heating rate and the final heating temperature. While most of the fractures were oriented more-or-less parallel to the bedding plane, some were strongly inclined relative to the bedding plane. The formation of inclined fractures is attributed primarily to the shape and orientation of a minority of the flake-like kerogen patches, and also to the effective stress field and fractured zones of weakness. A conceptual model is proposed to explain the dynamics of fluid expulsion and the associate fracturing behavior.

Numerical Modeling of Fracture Network Evolution in Organic‐Rich Shale With Rapid Internal Fluid Generation

“…Under low differential stress, however, our results show that inclined initial flaws can be essential to create fracture connectivity in the vertical direction, since external stresses do not cause rotation of propagating fractures in this direction. This mode of connectivity creation was previously hypothesized by Panahi et al (2018) but not investigated systematically. Therefore, if the local stress state is not strongly anisotropic, orientation of initial flaws on the microscale may be an important control on the fracture network geometry and thus fluid transport properties.…”

“…The characteristics of the phases of fracture network evolution as described in the interpretation section are in good agreement with results from previously published laboratory experiments. Based on time‐lapse synchrotron imaging of heated, immature shale samples (Kobchenko et al, 2011; Panahi et al, 2018), a three‐stage process model was proposed: (1) Fractures initiate from thin kerogen flakes or around elongated pores, then (2) propagate and coalesce, preferentially along lamination planes, but sometimes also obliquely, and (3) finally close under confinement after the hydrocarbons have been expulsed. A similar behavior was documented in analogue experiments analyzing CO 2 production in impermeable, elastic gelatin‐yeast‐sugar gels, though at a larger model scale, which is more comparable to our numerical setup.…”

“…However, our current understanding of hydrofracturing due to internal fluid generation stems primarily from laboratory models and a few numerical studies. Typically, the process observed in experiments can be subdivided into three phases, including (1) fracture initiation, (2) growth and coalescence, and (3) expulsion (Kobchenko et al, 2014;Panahi et al, 2018Panahi et al, , 2019.…”

“…Similar fracture initiation mechanisms were observed in analogue experiments using brittle, elastic gelatin‐yeast mixtures, where a drainage fracture network developed from elliptic gas bubbles (Kobchenko et al, 2014; Vega et al, 2018). Fracture propagation is thought to be driven by continuous liquid and gas production, and several factors may influence the propagation direction: (1) initial orientation of kerogen flakes or thin pores, for example, due to sedimentary lamination, (2) external stress field configuration, and (3) internal stress interactions (Kobchenko et al, 2011; Panahi et al, 2018; Vega & Kovscek, 2019; Vernik, 1994). Observations linked to fluid and gas expulsion in laboratory experiments include cyclic fluctuations of fracture apertures, that is, opening and closing, in response to fluid pressure buildup and relaxation due to fluid generation and burst‐like expulsion (Kobchenko et al, 2014; Panahi et al, 2018, 2019).…”

Layering in Shales Controls Microfracturing at the Onset of Primary Migration in Source Rocks

“…The kerogen patches within high organic content layers are modeled as penny‐shaped microfractures with a radius of 1 cm. Each microfracture extends over 6.5 discrete elements in the bedding direction of the shale material to reproduce the bedding‐parallel preferential orientation of kerogen patches observed using synchrotron X‐ray tomography imaging (Kobchenko et al, 2011; Panahi et al, 2013, 2018). The number of elements has been chosen as a compromise between a fine discretization of the layers, with 6.5 discrete elements across the layer thickness, and a reasonable simulation time.…”