Occasionally we see evidence for the diffusional path of hydrogen.
Many thanks to Luis Fernandes Silva for the photos and sample.
I was skimming through some photos that Luis sent me a few years ago when one in particular, from Sesimbra del Drento, caught my eye. This crag is a high-sulphate, sea cliff where the presence of sulphate reducing bacteria (SRB) causes hydrogen embrittlement of the stainless-steel anchors installed there.
Much of this blog has been dedicated to this complex failure phenomenon, so this short note is just an addition to stuff I have already discussed at length. Start here if you wish to catch up with the conversation.
Embrittled anchors are trivially snapped with a few swipes of a hammer.
This was the case for the Fixe A2 (304) anchors illustrated below. They had been in service for not much more than 10 years before removal.
But the fracture is a little weird, is it not?
Have a look at a selection of bolts from the same location.
The fracture surfaces labelled B are fairly uniform across the diameter of the bolt, whilst those labelled A show anomalous fractures toward the centre of the bolt. The other bolts exhibit differing degrees of non-uniformity of fracture.
I have no reason to believe that such poor material uniformity could exist at the time the bolts were installed. Surely, they would have been every bit as ductile as one would expect for 304 SS? My guess is that the lack of uniformity is to do with varying degrees of atomic hydrogen uptake during the years of exposure to the cliff environment. The initial point of attack by SRB will be the surface of the bolt in the low oxygen space against the side of the hole. Atomic hydrogen generated by cathodic processes at that surface diffuses in toward the centre of the bolt, and, for the case under discussion, it looks like hydrogen did not make it right to the centre. Thus, the outer layers are embrittled, but the core not so much, or perhaps, not at all.
I came across an analogous phenomenon some years ago at a time before it had occurred to me that we were dealing with a hydrogen embrittlement phenomenon. The photo below is the surface of a brittle-fractured bolt from Cabo da Roca.
I ground this surface down by 1mm or so and then polished it so I could examine the microstructure beyond the fracture zone. Having done that, on a whim, I dropped it into 50% nitric acid thinking it would immediately passivate. To my surprise, the outer material etched at quite a rapid rate leaving the inner core standing as an island.
The only conclusion I can draw is that the outer regions of the bolt are chemically altered compared with those toward the centre. We know that passivation by 50% nitric acid involves the formation of a protective layer of chromium oxide. Could it be that the difference lies in the presumably higher hydrogen content of the outer regions? Hydrogen could well suppress the facile oxidation of the chromium.
Yeah, nah … maybe. One thing for sure is that we see clear evidence of hydrogen damage in the microstructure of the outermost regions compared with the inside.
Sometimes I wonder if I’m becoming the guy that sees hydrogen embrittlement everywhere … bolt failure? HE for sure. Another failure? HE, why not? But no, I’d argue that such failures are rare simply because a whole bunch of circumstances need to align before the phenomenon manifests.
Firstly, you need to be reasonably close to a point on the earth’s surface where plate tectonics creates active volcanism. Secondly, there needs to be a marine transport path from the volcanic sulphur sources to your sea cliff. Thirdly, the environment at that cliff has to be one that sponsors the conversion of the elemental sulphur so transported to sulphate. Fourthly we need to attract sulphate reducing bacteria into anoxic regions between bolt and rock, and only then, do we have the right conditions to sponsor the cathodic charging of the bolt with hydrogen.
That is quite a long, causative chain. On the surface of things, it seems too fanciful to be real, but, with every passing year, evidence is accumulating that wherever we encounter a crag that eats 304 SS anchors, then this elaborate chain of events will exist.