I'll admit my knowledge of metallurgy is informed by a background in welding, but it's my understanding that the colors on that blade can only happen with a large temperature difference between the middle and its ends, likely as a result of the maker using a benzene torch on it for a few minutes. This high heat is going to do the same thing to the blade as it does during welding: it fucks up the temper. Heat treating is more than just making the metal hot; you have to make it uniformely hot, for a specific amount of time, and then cool it gradually and under control. Doing that doesn't give you the pretty colors, but it does give you stronger metal.

Rivalarrival

•

I won't say that this blade is properly heat treated; it probably isn't. In welding, the problem is the wide variation of heat affects in a very small zone. You can have material that is very brittle just millimeters away from material that is very soft and ductile.

You're describing "normalization", which is a process that makes steel uniformly tough, but "plastic". When you flex it, it bends, and stays bent. "Annealing" is a similar process, where the temperature is raised a bit higher, and the cooling slowed even more. "Annealing" leaves the steel very soft.

In tool making, you're first looking for high hardness (acquired with a "quenching" process). This makes it very brittle; it has no elasticity.

Next, you're dialing back that hardness with a "tempering" process, which is done at a lower temperature than the normalization process, and the cooling can be much faster. When tempered, it's still very hard, (significantly harder than "normalized") but now it is slightly elastic. It will flex, but beyond a critical point, it just snaps; it probably won't take on a permanent bend.

These colors are oxide layers that form at temperatures in the "tempering" range.

captain_aggravated

•

To expand on this, the rainbow of colors which start at a straw then turn yellow, red, brown and then that vivid blue, are caused by refraction. The oxide layer on the surface is transparent or translucent, and the thickness of the layer determines what wavelength of light it scatters. The hotter it gets, the thicker the oxide layer forms, so you can fairly reliably tell the temperature the metal has been heated to by eye, and you might use different amounts of heating to achieve hard-but-brittle or soft-but-tough.

I've even seen it done by Chris of Youtube channel Clickspring for decorative purposes. It's how he made the steel hardware of his brass clock blue.

Exactly how you temper something the size of a sword using a forge is a bit outside my understanding; I've done it with relatively small bits of drill rod to make lathe tools with a gas torch, but that's about it.

Rivalarrival

•

The hotter it gets, the thicker the oxide layer form

This is accurate enough for tempering of most cutting tools, but technically, the oxide layer will continue to grow if you hold a lower temperature for a longer than normal time, and might not fully develop if you reach a higher temperature for a shorter than normal period of time.

This property useful if you are trying to develop a specific color rather than achieve a specific metallurgy. You can heat to a lower temperature for a longer time to develop a deeper, more consistent color.

In my experience, it's easier to develop colors with an oven or propane torch rather than a forge or acetylene.

captain_aggravated

•

Yeah I referenced Clickspring, when oxidizing a part for decorative purposes he would put the part in a brass tray full of brass shavings apparently to function as a thermal mass so that the color comes out evenly.