Alloy embrittlement

[LyndonJ]

I just came upon this statement in another thread which i've heard before.

I was wondering if anyone else here had considered this.

Originally Posted by pistnbroke

...If you have ever tried to weld an old boat you will know that the aluminium becomes brittle and is not the material it was 40 years ago ....

Embrittlement of alloy over time in a marine environment definately seems to be an issue, the metal gurus say it's due to atmospheric chloride embrittlement and that alloy exposed for a decade has a much poorer fatigue response and develops a brittle surface. Not due to corrosion but to a change in the metal surface chemistry.

This suggests that for long life, alloy should always be coated. It also raises the question of significantly reduced fatigue response in uncoated material.

The crux is that uncoated alloy has a finite life. Something boat repairers are aware of but not necessarily designers and builders.

Should class societies require coatings ?

Does anyone have figures or papers on this ?

[Ad Hoc]

To give some meaning to this statement the answer tends to get metallurgical. And as such gets a bit messy..

However, aluminium is a ductile material, not brittle. It has an FCC structure and has 4 {111} slip planes, ie easy to move dislocations along their <110> slip directions. Hence being a ductile material.

There is only brittle failure, owing to manufacturing issues. Such as bending a plate into too tight a radius, the elongation, or movement is restricted under load; on the inner surface of the bend. This affects the behaviour and mode of fracture. The fracture, in this instance, is preceded by the formation of large plastic zones ahead of the crack or flaw or void. It would invariably be a crack, owing to the outer surface of the plate being over-strained. This would lead to a brittle failure along the inner bend surface, under fatigue loading, for example. But these are the exception rather than the rule.

So, ignoring poor manufacturing/QA as the cause for brittle failure, the alloy itself is ductile.

So for a failure to occur it is a combination of crack initiation, how much resistance there is to decohesion in the matrix, the amount of localised strain and grain size. Coupled with this the fracture toughness.

The “newer” aluminium alloys have greatly improved quality control in their manufacture. This has lead to improved fracture toughness. It has resulted in the control in the levels of impurities such as: iron and silicon, and also copper. Experiments have shown that plane strain fracture toughness may be doubled simply by maintaining these elements below 0.5% compared with alloys which have greater than 1.0%; more than likely these older alloys were in this ‘range’ i.e. greater than 1.0% in these impurities.

So the only link to such a statement I know of would be owing to the lower quality controls of such older manufactured alloys and hence an understanding of the fracture toughness that we know of today. Since the emphasises of aluminium development now focuses more on the behaviour of the alloy under varying conditions in service, rather than just the tensile strengths; such as 5383 etc.

So, in summary, alloys made/bought today are fine, fine for many many decades. Also, so long as they are fabricated correctly and with proper QA controls....the structure/vessel, shall out live you!

PS, forgot to add

For further reading try these, as starters:

Monodolfo, LF, “Aluminium Alloys: Structure and Properties”, 1976, Butterworths, London, UK,

Polmear IJ, “Light Alloys – Metallurgy of the Light Metals”, 3rd Ed., 2003, Butterworth Heinemann, ISBN 0 340 63207 0

Polmear IJ, “Aluminium Alloys – A Century of Age Hardening”, Materials Forum, Vol.28 2004, Institute of Materials Engineering Australasia Ltd.

Holroyd NJH, Vasudevan AK & Christoulou L, “Stress Corrosion of High-Strength Aluminium Alloys”, In: Treatise on Materials Science and Technology , Academic Press, 1989.

Court SA, Dudgeon HD & Ricks RA “Proceedings of the 4th International Conference on Aluminium Alloys”, Atlanta, USA, 1994.

PPS.
Finally, if the joint has been weld and rewelded several times, this seriously affacts the properties too. It reduces the mechanical and fatigue properties significantly as well as reducing its ductility. So, you need to know the history of a joint before passing judgement.

[kmorin]

LyndonJ,
without knowing the specific cases your references were dealing with I can't remark about those particular boats; but I can remark about some from first hand experience repairing, modifying and extending some hulls.

In late 50's and early 60's of the previous century Kenai Packers, an independent salmon cannery and packer in the mouth of the Kenai River, on the East Side of Cook Inlet bought three groups of 32' all welded aluminum alloy commercial salmon gillnet boats for use in Cook Inlet and Bristol Bay.

They were produced in Canada and the Puget Sound, and they're still in use in the Cook Inlet and Bristol Bay today (last season) although some of them are hard to recognize and some may have been turned to scrap as well?

I began welding on these 50-70 boats ( I don't know how many there were in all three groups- total) in 1977 and continued to work on various hulls until 1989 and even after that in 1995 I did another modification on one hull owned by a close friend.

During that time I added and removed tanks, decks, cabins, fishing gear, engines, keels, entire sterns and made more changes and repairs than I can recall without serious volumes of cold, fermented hops refreshment.

The original hulls were mainly 5052 but I don't know the aging or heat treatment grades for the plate and the main extrusions were 50 series in one group but 6061 and 6063 on the other two groups. All were MIG welded but I am not confirmed the filler wire was exclusively 5356 but that is what I used, exclusively, to MIG and TIG all the work our shop did.

Some hulls did have extensive and catastrophic exfoliation and others had incredible internal corrosion sites; both due to very easily explained reasons. There were some panels and formed pieces as well as some extrusions which clearly did not meet the standards for those alloys and shapes purchased by the builders. This is obvious since two extrusions, nearly side by side in the same application behaved differently over time.

If cleaned with a SS wire power brush after degreasing, the parent metal welded like new. There were extensive inclusions and deeply pocketed surfaces where an entire area had to be sanded down a few thousandths to get to 'weldable' metal but that was a cleanliness issue as far as I was concerned.

However, perhaps the layer you have been told existed was removed by our weld prep? and I don't realize that fact. I would never [willingly] attempt to weld on new plate that was not descaled and degreased so imagine our preconceptions when we had to work on a, then, 20 to 30 year old surface.

With a few hours under the hood as a comparison, once the oxide was off the underlying parent metal was just as useful as the sheet from the supplier today. None of the repairs cracked out, came loose or otherwise showed any signs of embrittlement, or any behavior different from new construction.

Aluminum's oxide layer makes the parent metal underneath chemically shielded and this "self healing property" is the sole reason there is a commercial metal product alloyed from this element. I seriously doubt the information's validity that the surface of material becomes changed enough to impact the parent metal properties deeper than the oxide: 3-4 mills.

I have also welded on these boats in somewhat cruder settings where cleaning wasn't always as meticulous as I'd have liked. Using a TIG torch, [AC w/ high freq] on old aluminum will still lift the surface coatings and vaporize them once they're broken up by the molten puddle below and flushed off the puddle with hot argon.

Even in cases like that, working on uncleaned old boat hulls I can't say I observed any difference in these boats. Dirty, a mess to weld (?) lots of sailor like terms about the mothers of the owners or welding helps: yes. Poor welds, weak metal, brittle or weakened 'old' aluminum: nope, I haven't seen that in these boats.

I suspect the basis for these claims comes from someone who has worked near or around a press braked section or an area of other post milling- forming, rolling or shearing. In those cases, there is definitely a possibility of strain hardening, that is well known, and over time those zones do show changes in properties. But as far as my experience goes, I've not seen these zones happen over an entire hull just from being exposed to salt air.

Cheers,
Kevin Morin

[CDK]

Of course the parent material doesn't change over time, only the surface does.
It is exactly the "self-healing" property of aluminum causing a change in composition and mechanical properties. The buildup of Al2O3 reduces the amount of Al at the surface, so the skin composition differs from the base material. The oxide itself is hard but extremely brittle: thermal expansion and vibration causes small cracks where new Al2O3 is formed.
A cross section of the material looks like a deep wrinkled skin under a microscope.

Furthermore in a marine environment not only Al2O3 is formed, but also complex salts with Cl and Ca that infiltrate much deeper than just a few mils.

[baeckmo]

Fully concur with Ad Hoc, KevinM and CDK here. Over the years, there has been circulating so much false information on aluminium in marine environment. Sometimes the lie is spread by someone who made a lousy job due to negligence/ignorance and trying to slip away from responsibility.

We have experience from commercial vessels since early 70-ies. There were a few examples of exfoliation, in particular where rolled edges were not cut away, and some plating we got from one Russian supplier. Some "surprizing" corrosion issues might be misinterpreted as "material change"; we have seen some corrosion where natural rubber has been attached to al in a wet, oxygen depleted zone . Probably the graphite in the rubber has caused galvanic corrosion.

But else, applying established cleaning procedures with welding, and avoiding known corrosion sources, al will stay intact in marine environment.

[LyndonJ]

Thanks for the responses.

I know it's only surface effects and from little defects big defects grow in higher stressed areas.

Originally Posted by CDK

.......
It is exactly the "self-healing" property of aluminum causing a change in composition and mechanical properties. .........
Furthermore in a marine environment not only Al2O3 is formed, but also complex salts with Cl and Ca that infiltrate much deeper than just a few mils.

This is more along the lines that I was interested in, samples of alloy tested after a decade of service life apparently had a much lower allowable SN (fatigue) cycle curve.

Of course I can't find the article now.

[Ad Hoc]

LyndonJ

You need to be careful here, since you're bordering on two different issues.

1) Fatigue and 2) SSC.

Fatigue of ally in sea water is dire!...really dire, this is nothing new, yet very few designers really appreciate that a losses of up to 90% of static mechanical properties can occur when ally is fabricated and in a sea water environment. You must also remember, ally has no fatigue limit too!

In the 6000 series graph, attached, you can see that a stress concentration factor of 12 (which is a serious defect), from a notch, has the same behaviour as a smooth sample in sea water!..that is really a major loss of fatigue life/strength.

SSC arises when the static stress and corrosion occur at the same time. Three conditions must be fulfilled simultaneously for this to occur, namely:

1. The alloy must be sensitive to stress corrosion (i.e. chemical composition, heat-treatment etc)
2. The environment must be corrosive
3. The duration and magnitude of stress must be sufficiently high

SSC, is attributed to an appearance of atomic hydrogen as a result of chemical reaction between Al and water, or of cathodic charging, followed by hydrogen diffusion into the crack tip. The crack of course comes from the fabrication and/or inservice conditions. The hydrogen present in the lattice is known to cause failure by decohesion of the lattice, decrease of flow stress in near - boundary area or by formation of brittle unstable hydrides.

SSC and fatigue are almost identicle in behaviour but are actually subtly different mechansims. It is a very complex mechanism involving metallurgical, mechanical and environmental parameters and results in brittle failure in alloys normally considered ductile, a si noted above. The fracture can occur along grain boundaries (intergranular) or through the grains, (transgranular).

[LyndonJ]

Yes thanks, I know that immersed fatigue is abysmal

I guess what I really want to know (and intuitively I feel that it's the case) is whether we can signifiacantly improve the characteristics by simply painting, ie keeping the choride ions away from the surface.

Long term fatigue from chloride surface intrusion-alloying causing stress accumulators that then crack oxidise and so on as CDK was saying, then the effect is similar to slip plane dislocations in bending but casued by localised 'surface' embrittlement.

so if we painted higher stress members eg vibration damp environments like engine mounts that might be a good idea.

The metalurgists seem to think so anyway.

I'd just never heard of this sort of concern.

[baeckmo]

Originally Posted by LyndonJ

........so if we painted higher stress members eg vibration damp environments like engine mounts that might be a good idea.

In those areas, plus some hard-to-reach places, where condensate with NaCl residues may accumulate, we have simply used ordinary corrosion-protective spray. Trademarks like Tectyl, Dinitrol etc. They are low-viscosity oil products that when dry have a waxlike surface and really stick to the al surface. If a weldjob should be required, it is easily wiped off with a nafta solvent. Much better than painting!

The building rules we have been using (DNV) for work boats, specify overdimensioned engine-room vents, causing more air than necessary for the engine, to circulate within the engine compartment, with the result of unproportionally great volumes of air moisture condensating on the free al surfaces. When asked about this, the authorities admit that the vent rules stem from GRP requirements; they didn't realize the consequences for al hulls!

[CDK]

I noticed several times already that a large part of this world's population is not familiar with these products (Tectyl, Dinitrol). In my opinion they are the best corrosion combat weapons on the market today.

Tectyl in several countries has a brand name popularity like Aspirin, Holland even made a verb for it. Its main application was aftermarket corrosion protection of the cavities in vehicle chassis, now applied by nearly all Asian and European car makers to enable them to promise up to 7 years of warranty against corrosion.

Dinitrol also does big business in the tanker world where it is used to protect the space between inner and outer hull.

[LyndonJ]

I'll look them up
Thanks guys.