Quantifying the influence of microstructure on effective conductivity and permeability: Virtual materials testing

• Published in: International journal of solids and structures Vol. 184; pp. 211 - 220
• Main Authors: Neumann, Matthias, Stenzel, Ole, Willot, François, Holzer, Lorenz, Schmidt, Volker
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.02.2020; Elsevier BV; Elsevier
• Series: Special Issue on Physics and Mechanics of Random Structures: From Morphology to Material Properties
• Subjects: Computer simulation; Conductivity; Constrictivity; Digital imaging; Effective conductivity; Engineering Sciences; Fuel cells; Materials; Materials testing; Mathematical models; Mean geodesic tortuosity; Microstructure; Morphology; Parameters; Permeability; Predictive simulation; Stochastic microstructure modeling; Tortuosity; Transport properties; Effective conductivity; Mean geodesic tortuosity; Stochastic microstructure modeling; Permeability; Constrictivity; Predictive simulation
• ISSN: 0020-7683; 1879-2146
• DOI: 10.1016/j.ijsolstr.2019.03.028

Effective conductivity and permeability of a versatile, graph-based model of random structures are investigated numerically. This model, originally introduced in Gaiselmann et al. (2014) allows one to simulate a wide class of realistic materials. In the present work, an extensive dataset of two-phase microstructures with wide-ranging morphological features is used to assess the relationship between microstructure and effective transport properties, which are computed using Fourier-based methods on digital images. Our main morphological descriptors are phase volume fractions, mean geodesic tortuosity, two “hydraulic radii” for characterizing the length scales of heterogeneities, and a “constrictivity” parameter that describes bottleneck effects. This additional parameter, usually not considered in homogenization theories, is an essential ingredient for predicting transport properties, as observed in Gaiselmann et al. (2014). We modify the formula originally developed in Stenzel et al. (2016) for predicting the effective conductivity and propose a formula for permeability. For the latter one, different geometrical definitions of the hydraulic radius are compared. Our predictions are validated using tomographic image data of fuel cells.