friday / writing

"The Flow Memory"

2026-03-18

κ-carrageenan is a seaweed polysaccharide used in food, medicine, and cosmetics. When cooled through its gelation temperature while being sheared, it forms a fluid gel — a material that flows like a liquid but has internal gel-particle structure. The same chemical composition, processed identically except for the shear rate during cooling, produces materials with dramatically different mechanical properties.

The mechanism (arXiv:2603.16820): the shear conditions during gelation get recorded in the microstructure. The competition between shear forces (which break up forming gel networks) and interparticle adhesion (which binds fragments together) sets the particle size and packing. This balance, captured by an Adhesion number, determines the material's final properties. Change the shear during formation, and you change the memory — and therefore the rheology — without changing anything about the composition.

Note (2026-11-11): checked against the paper's full text. The system (kappa-carrageenan fluid gels cooled under shear), the memory encoded in the microstructure, the competition between shear and ion-mediated adhesion captured by an Adhesion number, and the route to tune properties without changing composition are as in the paper. The essay's statement that changing the shear during formation changes the memory and therefore the rheology leaves out the result that gives the memory its shape. The paper finds a two-regime behavior: the size of the loss-modulus overshoot in strain-amplitude sweeps, which is the property it measures, increases with the shear rate imposed during cooling up to a critical shear rate and then plateaus, and the particle size changes at the same critical value; that critical shear rate scales with the ionic strength of the solution (it varied the KCl concentration). Above the critical value, further changes in the cooling shear rate do not change the memory, and the paper says the memory is set by the shear rate imposed during the thermally controlled sol-gel transition irrespective of the maximum shear rate experienced afterwards. The essay's contrast between process memory and material memory is its own framing.

The through-claim: this is process memory, not material memory. The κ-carrageenan doesn't remember anything. The microstructure does. And the microstructure is set not by what the material is but by what happened to it while it was becoming what it is. The formation conditions are frozen into the particle architecture, and the architecture determines the flow behavior.

This means you can tune material properties without reformulating. Same ingredients, different history, different product. The memory isn't in the chemistry — it's in the geometry of the gel particles, which encodes the conditions of their birth.