There was an interesting article in R&D Magazine, with a video:
.The mystery was always how the oil got to where it needed to be to affect metal removal - it was hard to see how the oil could penetrate into the solid metal that was soon to be removed. The video shows that the surface is cyclically sheared and ends up in a chip that resembles a stack of dimes, and no penetration into the bulk metal is required to lubricate this process.
Joe Gwinn