Engineered porous materials – ceramic filters, membranes, electrode structures, concrete, additive-manufactured components – derive their macroscopic function from microstructural features that conventional characterization methods cannot resolve. Permeability, transport selectivity, reaction efficiency, and mechanical durability all emerge from pore geometry, phase connectivity, and surface chemistry at the micron and nanometer scale. Bulk measurements average over these features and cannot predict material behavior from first principles.
Digital rock physics, developed over two decades in the petroleum industry, provides the methodology to bridge this gap. The same framework – multi-scale 3D imaging, image-based 3D model reconstruction, and physics-based pore-scale simulation – applies directly to engineered porous materials. We call this digital MATERIAL physics: the application of image-based, pore-scale computational methods to characterize, model, and predict the properties of non-geological porous materials.
Whether the material is a heap-leach ore, a shale reservoir rock, a ceramic membrane, or a battery electrode, the underlying challenge is the same: the macroscopic behavior emerges from microstructural features that continuum models treat as effective parameters. Those parameters cannot be reliably predicted from bulk measurements – they must be resolved directly at the scale where the physics operates. Our imaging and modeling workflow is material-agnostic; the pore-scale physics we resolve differ by application.
We apply our multi-scale correlative imaging and image-based modeling workflow to engineered and natural porous materials. 3D microstructure is captured directly by X-ray tomography and electron microscopy at the relevant scale, converted into a digital model using AI-assisted segmentation, and subjected to physics-based simulation to extract transport, flow, and structural properties directly from geometry – without empirical fitting of effective parameters.
The result is a predictive, image-grounded digital representation of the material: a digital material twin that links microstructure to function and enables design-driven optimization of porous material systems.
We characterize material microstructure across scales using a hierarchical imaging workflow: micro-CT/-XRM, SEM/EDS, FIB-SEM nanotomography, and S/TEM. Correlative registration across modalities produces unified multi-scale digital representations of the material.
Machine- and deep-learning segmentation and 3D model reconstruction for complex multi-phase material systems. Image-derived models become direct inputs to physics-based simulation.
Single- and multi-phase fluid flow, diffusive and convective transport, reactive transport, and structural mechanics on the digital material – extracting effective properties directly from resolved geometry.
Working on a porous material characterization challenge?
We welcome collaboration with manufacturers, research institutions, and national laboratories on digital material physics projects.