I'm not sure about this, but I'll throw it out for what ever it is worth.
I've been "playing with metal" for quite a few years. Some professionally and some...just for fun.
It seems to me that with regard to cylinder walls, what likely happens is a combination of two or more things. One, cast iron is relatively porous compared to steel. I learned when welding cast iron that if it was soaked with
oil, the
oil had to be burned out before welding. To me this means that the irons' porosity allows
oil to "soak in". Not at a molecular level...more like a sponge soakes up water.
If this is true, then as the rings seat and both the bore and the ring face get smoother (and seal), then some of the oil deposited is forced into the cast iron either as a burned by-product of combustion or as unburned oil.
Now, all things being equal, (we need a metallurgist here) carbon impregnation or not, sufficient temperatures have not been reached to alter the molecular structure of the iron. Neither has a "chill factor" been involved that I can see. (The chilling process, makes the iron extremely hard and dense.)
If any of this blather is acceptable then, I submit that what IS occurring, is a "filling" process. Iron particles combined with carbon are "glazing" the cylinder bore and providing a more contiguous surface and therefore, better sealing.
Ok....blow me away if I'm off target....
Al