Sauce at the Interface

Structural question — when does an interface layer change from binder to lubricant?

The same material can hold an assembly together and take it apart. Which one it does is largely a question of how much of it is present.

Bond, film, bearing#

An interface layer does structural work only while it is in contact with both faces it joins. Thin enough, it fills the surface irregularities of each and resists shear by adhering to both, and the two layers then move as one. This is the useful regime, and it is narrower than it appears.

Add material and the layer stops transmitting force between the faces and begins carrying it internally. The limit becomes cohesion within the layer rather than adhesion to the faces, and cohesion in these materials is low. Add more again and the film exceeds the roughness of the surfaces: the faces no longer touch anywhere, and the only resistance to sliding is viscous drag, which rises with the speed the layers are already moving at and falls as the film thickens. A resistance that appears only once motion has begun is not a bond. It is a bearing.

Tribology calls this progression the move from boundary to hydrodynamic lubrication. The course borrows the term deliberately and approximately — the pressures, speeds and materials here are not those of a journal bearing — but the shape of the behaviour transfers, and borrowing it lets you predict the transition instead of merely reporting it afterwards. The transition is continuous in the film and abrupt in the assembly: nothing visible changes until applied shear exceeds what the film can carry, and then the whole displacement happens at once, because a film that has started to slide resists less than one that has not. Interface failures are therefore recorded as events, with a time, rather than as trends.

Investigation#

At the sequential incident recorder, reconstruct one shear event from three frames and diagnose where it occurred. The diagnostic is residue. Material left distributed across both faces indicates the film failed within itself — cohesive failure. Material clinging almost entirely to one face indicates the bond to the other face failed first — adhesive failure, at a face you can name. Scrub between the frames and fix the frame in which movement begins.

Failure condition#

The upper layer displaces as a single event, with no measurable movement in the preceding frame. Record which face retained the film.

What leaves the session#

An Interface Incident Report: three sequential frames of one shear event with their interval, a stated failure mode — adhesive at a named face, or cohesive within the film — the evidence in the frames supporting it, and one alternative reading you considered and rejected. The Component Test falls due at the end of this week, and a rejected alternative is exactly the form its interpretation and limitations section takes.