A critical analysis of shale laboratory permeability evolution data

Highlights: • The boundaries of permeability ratios (k/k₀) are delineated by a generic model. • The boundaries of k/k₀ are determined by shale microstructural characteristics. • The effects of stress, sorption and slippage on k/k₀ cannot be separated simply. • High-level knowledge of sample structure and process interactions are important. This review study aims to identify why current experimental measurements of shale permeability are not consistent with predictions of commonly-used permeability models and explore how the identified reason would guide future research. These goals are achieved through the collection of experimental permeability measured under constant confining pressure (CCP) and constant effective stress (CES) conditions. These data show that permeability ratios (k/k₀) vary between an upper bound and a lower bound. A generic permeability model is developed to delineate the boundaries of k/k₀ based on shale microstructural characteristics. It's found that for upper bounds are controlled primarily by the fracture behavior while for lower bounds are controlled by the matrix or intact shale behavior. These findings suggest that the model predictions represent only behaviors of either fracture system or matrix while the experimental measurements from CCP and CES observations are for real shales. For real shales the internal dependencies among these factors cannot be fully understood through the nature of the external boundary conditions in CCP and CES tests, alone, but require high-level knowledge of sample structure and process interactions. The nature of permeability time dependencies on the internal process interactions must be reflected in any future experimental and modeling research.