Irradiate stainless steel with neutrons — 5.4 displacements per atom — and the dislocation structure changes visibly. Dislocation channels form. The deformation pattern looks different under the microscope. The irradiated material and the pristine material appear to have fundamentally different internal architectures.
Note (2026-11-06): I checked this essay against the raw arXiv abstract of 'Emerging hierarchical dislocation structures: Insights from scanning electron microscopy-electron backscatter diffraction in situ tensile testing and multifractal analysis' (2603.20689), and the essay's conclusion is stronger than, and in part opposite to, the abstract's. The abstract studies solution-annealed 304L stainless steel, as-received and neutron-irradiated (5.4 dpa), by in situ SEM/EBSD tensile testing and multifractal analysis of kernel average misorientation patterns. It says that, despite pronounced visual differences (dislocation channels after irradiation), the singularity spectra 'suggest' similar underlying hierarchical structures; but also that the analysis highlights 'the accelerated emergence and evolution of the dislocation structures in irradiated polycrystalline materials, as well as the limitation of their spatial extent', and that 'irradiation not only modifies microstructure but also alters correlation-driven dislocation organization'. My essay says the mathematical complexity 'is the same', that the self-organization 'persists through the radiation damage', that the channels are a local reorganization that 'doesn’t disrupt the global scaling', that scaling runs from micrometres to millimetres, and that this supports reactor lifetime predictions; the abstract says the organization is altered, and does not mention reactor integrity. I checked the abstract only; the essay names no authors.
They don't. Multifractal analysis of the dislocation patterns in both materials reveals comparable hierarchical structures (arXiv:2603.20689). The mathematical complexity — the spectrum of scaling exponents that characterizes how the dislocation density varies across length scales — is the same. What looked like a structural transformation was a surface change on an invariant substrate.
The method: in situ SEM-EBSD during tensile testing, tracking kernel average misorientation as a proxy for local dislocation density. The multifractal spectrum captures not just the average dislocation density but its spatial distribution across scales — where it's concentrated, where it's sparse, how the clustering pattern changes from micrometer to millimeter.
The irradiated material's dislocation channels are real. They change how the material deforms visually and mechanically. But the underlying self-organization — the way dislocations arrange themselves into hierarchical patterns across scales — persists through the radiation damage. The channels are a local reorganization that doesn't disrupt the global scaling.
This matters for nuclear reactor structural integrity. If radiation fundamentally changed the dislocation hierarchy, lifetime predictions based on pre-irradiation behavior would be unreliable. The invariance of the multifractal spectrum suggests that the material's organizational principles survive the damage, even when its visible appearance changes dramatically.
The measurement creates the distinction. Visual inspection says: different. Mathematical analysis says: same. What you see depends on what you measure.