S.A.I.L. = Sixth Attempt In Learning

Hello there, dear reader, and a happy 2024 to you. For aerospace engineers, like yours truly, that number has a special place in our hearts: it’s the designation of the most commonly used aluminium alloy in airplanes. With that factoid in mind, I thought I’d treat you to my sixth material failure lesson, which also has an aluminium flavour to it. And it starts with this question:

?

“What if we could make aluminium as strong as titanium?”

?

At first sight, this seems like an impossible dream. Aluminium alloy 2024 has a yield strength of a mere 345 MPa, meaning it will begin to deform plastically if stressed beyond that level – and that’s assuming you’ve heat treated it to the “T6” temper. By comparison, titanium alloys typically do not yield before the stress exceeds 1,050 MPa or so. Quite a difference! Now, 2024 isn’t the strongest of the lot: that honour goes to an aluminium alloy with the unimaginative designation of 7068, which yields at 683 MPa. Way better than 2024, but definitely not in titanium territory.

?

Now, time for a little history. In 1976, Dutch material scientists Ton Bendijk and Laurens Katgerman were experimenting with “rapid solidification”. They melted aluminium and pored it onto the outer surface of a huge, spinning copper wheel, water-cooled for extra oomph. This way, they managed to solidify the molten metal in well under a microsecond, translating to a cooling rate of a million (!) degrees per second. I don’t know about you, but to me this certainly qualifies as “rapid”.

?

Ton and Laurens went to all this trouble in the hope of creating a fully amorphous metal – so, an alloy without the regular atomic ordering we see in all metals. That didn’t work, but they did find something interesting: the thin strips of aluminium that got shot off the copper wheel were exceptionally strong. Metallurgical analysis soon revealed the secret: thanks to that ultra-fast cooling, the metal’s micro-structure was very, very fine-grained, and that’s a well-known strength booster – look up “Hall-Petch relation” if you’d like to know more.

?

And Ton and Laurens didn’t stop there…

?

With rapid solidification, they reasoned, you can create alloys that can’t exist otherwise. Take those 7000-series alloys, like the aforementioned 7068 and the more common 7075 – another aerospace favourite, by the way. Basically, they get their strength from the zinc that gets added to the aluminium. Now, if you add too much of it, then the zinc will go its own way, Fleetwood Mac-style, and “segregate” into a separate phase, doing more harm than good. But with rapid solidification, you can simply flash-freeze the zinc into place, creating what metallurgists are fond of calling a “meta-stable microstructure”. Could those titanium dreams really come true?

?

By the way, if you’ve never heard of rapid solidification alloys, don’t panic: neither had I, until I joined the Techno Center of Royal Nedschroef, all the way back in 2005. FYI, Nedschroef is a manufacturer of fasteners – think nuts and bolts – working mainly for the automotive industry. You’ve probably never heard of them, but chances are that you have some of their products in your car. They already had a line of aluminium products, all made from alloy 6056, and they were hoping to expand these “Nedlite” bolts with something stronger, made from those “RS alloys”. They already had a project going, working with a small Dutch company specializing in making the bulk material, and some initial steps towards these "Nedhilite" bolts had already been made.

?

That’s where I came in.

?

Long story short: did I get anywhere near titanium-strength? Well, at one point I tested a batch of RS aluminium bolts that got to around 800 MPa before yielding, which was the target value at the time. That’s impressive in its own right, nicely beating the 7068 alloy. But there was a lot of scatter in the results I got, with many bolts falling well short of that target.

?

And, there was another problem. Those RS alloys were still made with a big copper wheel, meaning that their first solid form was one of long, thin strips that Ton and Laurens would immediately recognize. In that form, the alloy was of course supremely suited for exactly nothing, so the next step was to stuff enough strips in a cylindrical steel container and compress them into a kind of porous briquette. To get rid of the porosity, those briquettes were then HIP-ped, which was short for “hot isostatic pressing”. Unfortunately HIP-ping is slow and expensive. I ran some numbers, and while the 800 MPa strength was by itself super-nice, the cost price was simply way, way over anyone’s budget.

?

Then I hit a Catch-22. The only way to speed things up was to increase the temperature during HIP-ping, but that would give more mobility to those pesky zinc atoms, giving them a chance to segregate. Meta-stable indeed! In other words, I could cook my goose slowly to get good strength at high cost, or crank up the heat to reduce cost but at the price of the exact property I was after.

?

I decided it was time to talk to the experts, and called some old friends at Alusuisse.

?

They knew about RS alloys (of course), and needed only a brief introduction to my goose before breaking the bad news: this particular Catch-22, they solemnly declared, was not an engineering problem but a physical one. And, they had more bad news. You see, those HIP-ped briquettes weren’t suitable input for our bolt makers. So, they were drawn into thick wire, and that form was okay. However, this drawing operation is really hard to do well, and some inhomogeneity in the output is unavoidable. Hence the scatter I saw in my strength tests. Now that was in fact an engineering challenge, and one that Alusuisse would have been happy to tackle with me. If all went well, the resulting fasteners would still be way too expensive for the car industry, but in aircraft they stood a damned good chance: kilo for kilo, those RS aluminium bolts would outperform titanium, AND be competitive at that. Small wonder, because as true aerospace geeks know, titanium bolts are obscenely expensive!

?

Nedschroef management decided against the proposal, and that was that. Ton, Laurens, I’m sorry: this was one I just couldn’t get off the ground, or indeed, on the road.

?

What I learned from this material failure is, frankly, as simple as it is profound: test bars are one thing – actual products are something else entirely. The processes used to bring your material into the desired shape can, and will, affect the properties you are after, and usually in a bad way. Especially with metals! Let this be fair warning for all material enthusiasts out there.