re-heat treating old knife blades

If you are saying that the fracture is running along the grain boundaries of large grains it is not hydrogen related. It is some form of grain boundary embrittlement. Could be overheating / burning, or blue brittleness, P levels etc.

Cool, thanks for the response. :)

Well we mounted and polished a couple broken-off blade tips etched with nital and the grain size seems to be about 9. Metallurgy teacher thinks he sees iron oxide pockets in them. He says it's simply me over heating them.

I'm not going to disagree with that.

But with me having pretty good luck re-heat treating my own heat treated 1095 (that hadn't been used, they are just too warped so had another go at it;) and having had pretty good luck with re-heat treating an old butcher knife last week, that hadn't been used in a kitchen...

I'm wondering about the old knives having soaked up hydrogen and/or oxygen while in food service.

What do you guys think of that? Possible?

P levels, although unknown in the knife blades, are known in the old files, which are low in P and can be just as bad at getting "crumbly".

I'm going to re-heat treat a couple particularly rusty old files and knife blades as an experiment.

I'm not really setup to anneal. :/

What about the idea of directly austenitizing martensite tho? Anyone got any ideas about problems with that?

Haven't really found anything written addressing austenitizing martensite directly.

Tried normalizing old knifeblades and files and that seemed to make them worse instead of better. :/

I've been working on this problem for 15(?) years. :)

I'm getting to the idea that it's me and my sloppy heat treating methods (blacksmith style, I use the "arrest point method" instead of the "magnet method") that along with other minor problems, sometimes they gang-up and ruin the blade? :)

I'm not anywhere near giving up on this yet.

Alvin in AZ

I would look for oxides in an unetched sample, they are easier to see then. You need to have a very clean well polished surface as staining from cleaning can be confused with small oxides. The burnt samples I have seen, sometimes show oxide penetrating along grain boundaries near the surface.

Hydrogen can cause problems in service for very hard materials. The commonest source of atomic hydrogen is from corrosion. Rusting on springs or high strength bolts can cause hydrogen embrittlement while the corrosion is taking place. During food use unlikely to be an issue. Oxygen is not an issue unless the part is far hotter than it should be for normalizing. Definately not an issue at room temperature.

Martensite is generally a homogenous structure so it is generally a good starting point for austenitising as there are no large carbides that would be difficult to bring into solution.