One issue is that if you create surfaces that are too smooth, they can end up fusing together; especially if they are made of the same material. Cold welding is one term for the effect.
In many bearings the two surfaces don’t touch at all, but are kept apart when running by a continuous film of lubricant; this is hydrodynamic lubrication. This extremely widely used, for example in car engines, and means bearing wear is almost non-existent. (it does occur when starting up)
Except that then, instead of wear on the solid surfaces, you have wear on the liquid. And probably a lot more of it: Every time the machine is used, you’ll lose a little bit of lube, and need to replace it every so often.
Wear on the liquid? If you mean that oil in bearings undergoes chemical change, it does not.
And you don’t lose any oil because it is continuously recycled in the engine.
You do lose some, as evidenced by the fact that you eventually need to replace it.
The von Mises criteria is a way to evaluate the root square sum of all stress within a finite volume of solid material treated as a continuum, and as such is a scalar property. It isn’t very useful for estimating the stress at an interface because of the discontinuity of the continuum. Normal contact stresses are estimated using the Hertzian contact stress, while estimation of shear stress at the interface is bound up in the coefficient of friction and the hardness, both of which are measured instead of fundamental properties.
In general, it isn’t particularly meaningful to talk about the stress at an interface; in the case of a solid/fluid interface the quality being considered is the aerodynamic or hydrodynamic drag, or the distributed shear force applied to each continuua at the interface. For solid-to-solid interfaces that are sliding with respect to one another the evaluation is more complicated because of the variety of different mechanisms and changes to the surface that progress through the interaction.
In general (and in response to the question of the o.p.) when two solid surfaces are sliding with some non-zero contact pressure and no lubrication layer, there will always be some degree of deformation of one or both materials which will at least remove the oxidation layer and cause microscopic gouges and dislocations because within the contact area there will be local ‘high’ spots where the normal stress is actually very large (because the force is concentrated), and when they wear away there will be another high spot, et cetera. With two very smooth and flat surfaces, like panes of glass, they will actually form a partial vacuum, effectively increasing the normal force and the effective coefficient of friction will go way up.
This is not correct. In general, fluids don’t ‘wear’ or radically change in character due to stress unless the internal viscous stresses become so large that the waste heat generated causes the fluid to break down chemically or become ionized. The reasons that lubrication needs to be added or replaced is because volatile components evaporate under time or heating, and because of contamination where surfaces come into contact and generate microscopic debris. In automotive engine oil, for instance, the aromatics heavy petroleum distillate will evaporate (there is some thermal breakdown from engine heat as well, but even oil sitting in an unused car will lose the lighter constituents and become more viscous) and particles from the engine block and piston rings will accumulate. Engines also may lose some amount of oil past the pistons or through seals on the sump; generally a very small amount with a well-built and properly maintained engine, but that isn’t because the oil is ‘wearing’.
Stranger
In what sense is loss of material at the surface not “wear”? Sure, it’s a different mechanism for a liquid than for a solid, but it’s still loss of material at the surface due to relative motion at that surface.
I do not understand the point you are trying to make. The movement of the lubrication oil is not “wear” in any sense of the word, and it is not consumed in the process of providing a hydrodynamic layer other than the typically very small amount of oil that incidentally leaks past piston rings and is burned in the engine.
Stranger