F.A.I.L. = Fifth Attempt In Learning

“Recyclability is NOT a material property”.

?

Welcome, dear reader, or welcome back. Last time, I hit you with lots of text and then I didn’t even spell out what I’d learned from my material failure. In this fifth installment I decided to start with the lesson, so there it is! You're free to go off and celebrate the New Year – unless you’d like to learn where this nugget of material wisdom comes from, in which case I invite you to stay. It’s a lesson that applies to all sorts of things, and that includes windmills, like the ones pictured above as clickbait.

Thirty years ago to this day, I was busy doing my MSc graduation work, securely sequestered in the Structures & Materials Laboratory of the TU Delft. The focus was on something called “GLARE”, or “glass-reinforced aluminium laminate”. It’s a kind of material lasagna of thin layers of glass-epoxy composite (think 0.25 mm) alternated with aluminium sheet metal of comparable thickness. This tasty combo offered the kind of properties to make any aerospace engineer drool with anticipation, combining superb fatigue strength with excellent impact resistance and outstanding corrosion behaviour – in short, it could withstand the unholy trinity of dangers awaiting any material destined for primary aircraft structures. All very nice and tidy I thought, but can we actually recycle this stuff?

As it turned out, it wasn’t too difficult to separate the different constituents of this new aerospace material. In fact I found four solutions, with the most successful one also being the most spectacular: cryogenic separation. In plain English, I’d collect a bunch of GLARE waste parts (easy to find in a lab devoted to destroying samples in all sort of creative ways), cool them down with liquid nitrogen, chuck it into a shredder, and collect the output. Sounds complex, but really quite easy, and to my happy surprise, liquid nitrogen wasn’t expensive at all, it being a by-product of production of the stuff everyone in heavy industry wants all the time, LOX – or “liquid oxygen” to non-rocketeers.

Following this violent but efficient process, the next step was to separate the metal from the composite flakes. For that I used a device called an eddy current sorter, a.k.a. “aluminium magnet”. That yielded virtually pure aluminium flakes in one box and equally-pure glass-epoxy composite in the other, and from there I supposed industry could recycle these “fractions” in their usual way. Problem solved!

?

Or was it?

?

The real question, I discovered, was one of economics. Recycling is done to make money, and if you can’t break even when processing a particular waste stream into new materials, then that stream is de facto not recyclable. Now that cooling with liquid nitrogen, shredding, and eddy current sorting weren’t that expensive in a lab test, but in the real world you’d need to make a proper investment to get things going. To get anywhere near positive in terms of ROI (= return on investment, glad you asked) I calculated you’d need at least one ton of scrap GLARE per hour. In the usual two-shift production, that’s around 80 tons per week. By comparison, the main aircraft using GLARE today is the mighty Airbus A380, with 8.3 tons of the stuff in its structure, all in locations where the other materials just won’t do the trick. So, you’d need to recycle nine A380’s per week to break even, in terms of that new material. Quite a glaring (sic) difference with how many of these planes actually got built. And I haven’t even mentioned the cost of collection.

During my subsequent PhD (I know, “piled higher and deeper”, guilty as charged) I worked with all sorts of composites and laminates. It’s a colourful list: glass-ULTEM composites from Ten Cate, carbon-PA12 from a German start-up, stuff called Xenoy and Valox from GE Plastics (used e.g. in the BMW Z1), aluminium-foam sandwich plates from Fokker Special Products, and quite a few more. And for all of them, I found feasible recycling solutions, provided you can collect enough of it to make a dedicated recycling plant run in black figures. A key feature of all these “advanced materials” is that kilo for kilo, they are all relatively expensive – 10 euro/kg or more. However, there was simply no way to preserve that value during recycling, as that inevitably breaks up the carefully-optimized composition that is often the primary source of value in the first place. That’s very different from, say, titanium, which is also expensive but which can be recycled without too much loss in value.

?

The operative word is, of course, “enough”.

?

Let’s say you’re Volkswagen, and you’re building a super-economical small car. Let’s say that instead of plain aluminium, you use a bit of some kind of aluminium-plastic sandwich sheet – perhaps as much as 20 kg per car. And (big assumption!) you’re successful, selling 10,000 units per year across the EU for eight years’ time. Total volume of sandwich sheet: 1,600 tons. Sounds like a lot if you’re used to lab scale work, but in the real world it’s peanuts. A single recycling plant, typically eating way more than a measly one ton per hour, can process that much in a few weeks – but how are you going to collect all of that material, seeing that it got spread out Einstein-style “across space and time”? Really, it’s more economical to just chuck those cars into the regular car shredders and accept that the laminate isn’t recycled nearly as well.

FYI, Volkswagen actually did something like this, with their 3L Lupo – a “special edition” of the regular Lupo subcompact, produced from 1998-2006. It was indeed super-economical, consuming just three litres of fuel in 100 km (hence the name), and it did in fact use a bit of aluminium-plastic sandwich sheet called “Hylite”, though nowhere near 20 kg/car. Like GLARE, it could have been recycled via the cryogenic route, but in practice this of course never happened.

?

So, to reiterate the lesson: recyclability just isn’t a material property. It all depends on how much you apply, and how, and where. Much like reliability, in fact, which also cannot be traced back to materials alone. Case closed! What we CAN say is how much of material’s current supply has been used before, i.e. is recycled, but that is purely an economic property of our supply base. It’s a very different conclusion from what I started with back in 1993, and one I am very happy to have learned. Glorious failure strikes again!

Incidentally, you may wonder if should use such composites, laminates and other “monstrous hybrids” at all. That question I also tackled during my PhD and the answer is, plain and simple, YES - in selected cases. Just make the LCA’s and you’ll discover that weight savings can be incredibly successful in preventing total life cycle emissions, in cars, trucks, and especially airplanes. I’ll leave a link below to one of my own LCA studies, nicely peer-reviewed. As for windmills, I’ve recently learned that yes, those glass fibre composite blades are indeed very challenging to recycle – but after just seven months, they have generated more energy that what went into producing their materials, and then they’re still good for the next twenty years or so. Piles of discarded blades might look bad by themselves, but in the grand scheme of things it’s a trivial problem, no matter what Big Oil propaganda may want you to believe.

?

Five down, five more to go. But that’ll be in 2024. Hope to see you then.

?

?