Re:A 3K-year-old copper smelting site could be key to unders...
by Sharlin on Thursday October 02, 2025 @11:02AM
(#45450594)
Thanks, I was a bit confused by the different allotropes. Interesting that the martensite transform naturally only happens on the surface (which as you say is actually desirable) due to the rapid cooling it requires, I didn't know that. But of course it makes sense that quenching doesn't instantly cool the bulk of the workpiece.
Austenite is one of the high temperature crystal structures of iron, and is not usually seen at room temperature except in certain non-heat-treatable stainless steels, and in highly alloyed steels some small amounts of retained austenite remains that doesn't get the chance to transform due to low cooling rates and suppressed M_f temperatures. Martensite forms by rapidly cooling austenite without giving the atoms the ability to re-organize into their preferred structure (ferrite, pearlite, or cementite depending on the carbon concentration). This transformation actually changes the size of the crystal matrix, which locks in a ton of internal stress as atoms want to move around but can't. All of this internal stress must then be overcome by an external stress to move the atoms around, resulting in a much harder material.
Martensite requires extremely fast cooling rates (on the order of 100s of degrees/second), which is why most carbon steels are quenched to harden them. These cooling rates are only able to be achieved a little ways into the bulk of the material, so you usually end up with a hardened case made of martensite, and a softer more ductile core that is usually pearlite (layers of ferrite [pure iron] and cementite [iron carbide]) that form due to the slower cooling in the core. This is usually actually more desirable than an entirely through hardened piece, as the hard surface can resist wear and indentation, while the soft core increases the ductility and toughness which reduces the risk of fracture.