Materials & Energy
Equal Porosity, Different Diffusion: Test the Hidden Connectivity
2026-09-10
Two digital materials can have the same porosity and very different modeled transport. An intuition about connected pathways becomes testable by holding void fraction constant while varying geometry. This can identify a useful descriptor before a materials team relies on a single aggregate number to rank candidates.
The computational starting point
Start with licensed existing images or synthetic binary volumes for a non-hazardous transport problem. OpenPNM provides an effective-diffusivity example with explicit conventions. Check voxel scale, segmentation and boundary connectivity. A small cropped image may disconnect a pathway that continues outside the observed volume, changing the meaning of the calculation.
Sources: OpenPNM effective diffusivity example.
Where intuition enters
The candidate explanation is that a few narrow connections dominate the accessible routes. Define the expected effect on directional effective diffusivity. The rival is a segmentation or finite-volume artifact. A visually tortuous image does not establish which of these explanations controls the computed result.
A test that can disagree
Generate matched-porosity structures and compare directional transport under identical boundary conditions. Use an image-based solver such as PoreSpy for a complementary calculation where appropriate. Agreement between tools is useful only after checking that both solve compatible equations with compatible conventions, not merely that their output labels match.
Vary segmentation thresholds, domain size and resolution within a frozen sensitivity set. Include a straight-channel reference and a disconnected control. If the ranking reverses under a plausible one-voxel boundary change, report structural uncertainty. Do not convert a modeled diffusion coefficient into a measured permeability or a broader material-performance claim.
Sources: PoreSpy finite-difference tortuosity.
An illustrative decision
Imagine two candidate images with equal void fraction, but one contains a narrow bridge that carries most of the simulated flux. Removing that bridge within segmentation uncertainty changes the ranking. The valuable output is a warning that connectivity evidence, not additional optimization, currently limits the choice between candidates.
What the research would deliver
A client would receive a transport comparison, descriptor shortlist and uncertainty map. This can guide a computational materials screening decision without requiring fabrication or exposure to chemicals. The preview identifies whether connectivity is a promising discriminator; detailed model validation and application-specific performance claims need a separate evidence scope.
Questions this raises
Is tortuosity defined identically everywhere?
No. Geometric and transport definitions differ. The report must state the convention used before comparing numbers.
Would matching porosity isolate every other factor?
No. It controls one descriptor. Domain size, connectivity, surface assumptions and boundary conditions still need explicit treatment.
Sources and their limits
- OpenPNM effective diffusivity example. A pore-network transport example and its conventions.
- PoreSpy finite-difference tortuosity. A digital-image transport solver, not a complete material characterization.
Prepared with AI assistance. The linked sources support the specified technical points; they do not validate applied psionics as a whole or guarantee a result for a client.
Read the editorial and evidence standard.
Continue reading
Explore Scientific Oracle consultingfor a scoped review of an existing-data research decision. Start with a non-confidential outline of the question, available evidence and the decision it needs to inform.