Scandium’s impact on the Additive Manufacturing of aluminium alloys

In metallurgical terms, scandium forms a precipitate in solution with aluminium, but we digress. It is a solution because the Additive Manufacturing of aluminium alloys is facing ever more demanding requirements – both technical and economic. The future of AM is in part manufacture, so innovation will increasingly be focused on how to reduce part cost and improve performance. The AM machine is an enabler: it enables new designs. Materials come into the frame to bring out the performance of the design. A lighter, lower-cost solution is achieved when these factors work in harmony. We both love applications, design, and materials and the amazing intersection that can emanate from them. We also love scandium and believe it is a solution for aluminium-based parts produced by AM.

We have written previously on the virtues of scandium-containing alloys of aluminium, specifically Scalmalloy® [1]. In that article, we generated a trade study looking at the overall part cost of an aluminium brake caliper traditionally manufactured from machined 6061, additively manufactured from AlSi10Mg, Scalmalloy, and Ti6Al4V. That study asserted that Scalmalloy, despite the powder being three times higher in price versus AlSi10Mg, resulted in a significantly lower cost AM part.

This may seem to be an unexpected result but, due to the high cost and low productivity of AM machines, the cost of AM parts is heavily driven by the build time on the machine. This means that any weight saving in the part which can be achieved due to improved material properties, results in a significant reduction in the production cost. In AM, weight is time and time is money. Raw material cost, which is a major input cost in traditional manufacturing, takes a back seat concerning overall part cost. This result exemplifies the role of design and material properties in the economics of Additive Manufacturing.

Today, aluminium AM represents a small percentage of the overall metal AM market, let alone the aluminium market. This is in part due to the current material offerings for the AM of aluminium compared to the marketplace of existing aluminium applications. Current applications made from cast or wrought aluminium start with raw materials priced from $2-10/kg, and which are easily machined with high rates of material removal and relatively low tool wear.

This economic aspect limits the niche of applications where AM makes sense in aluminium to those geometries which cannot be machined and are in low volumes where die-casting doesn’t make sense.

Or does it? With a high-strength aluminium alloy, you can additively manufacture something which cannot be easily achieved by machining or casting, at any volume of production. This is the combination of high mechanical properties with the complexity of casting-type geometries. In applications where this combination of capabilities is valued, high-volume production of aluminium parts using AM can be viable.

Fig. 2 shows a Venn diagram, which highlights this niche. In the diagram, this may look like a very small niche, but in terms of economic volume, the market size for serial production of such parts is orders of magnitude bigger than the market for prototypes in aluminium. This is because the potential volume of parts to be produced in this niche is much higher, and we are talking about serial production, albeit of a small number of different parts where these properties are important.

Additively manufactured AlSi10Mg is, at best, a cast aluminium replacement in strength-driven applications, and high-strength aluminium alloys often exhibit hot cracking when built because they are not weldable. Specifically, the alloying additions of the high-strength 7000 series alloys derive their strength from elemental additions, some of which volatilise during the build process.

So, what’s an aluminium metallurgist to do? The answer is to go with a higher-strength alloy that is also weldable. Scalmalloy is one such material. But if it solves the problem, why don’t more people use it? Cost is the first response, but we’ve already considered as to why that may not be true. Perceived scarcity or supply risks related to scandium are the next perceptions people raise.

What is scandium and where does it come from?

Mendeleev predicted the presence of scandium in 1869. Lars Frederik Nilson detected scandium in 1879 and produced 2 g of scandium oxide. The new element was named scandium, which is Latin for Scandinavia [2]. Metallic scandium was first produced in 1937 but the first pound of 99% pure scandium wasn’t produced until 1960. Russian and US patents were issued in 1969 and 1971 respectively, and most of the development occurred initially in the former USSR and the US.

Scandium is 21 on the periodic table, nestling in between calcium and titanium. It is classified as a rare earth element, which of course doesn’t mean it is rare in the scarcity sense. Scandium is the 32nd most abundant element in the Earth’s crust, just slightly more abundant than the lithium we rely on for many modern batteries [3].

Scandium is fairly abundant globally, including Canada, Australia, the Philippines, Norway, Madagascar (we could go on), but tends to be very thinly spread; that is the challenge of producing it. You need to process a lot of mining tailings to extract the scandium which is there, as it is present in low concentrations.

It is the cost economics of extraction that has led to the majority of scandium being produced in China rather than any special availability of the ore there. If you want to find the richest source of scandium to refine, you should head to Madagascar. The richer resources of scandium, such as those in Madagascar, are currently not refined, because it is cheaper to process the tailings of other mining activities to service the relatively small market demand that currently exists.

Today, only 15,000 to 20,000 kg per year of Sc₂O₃ are produced, with demand slightly exceeding supply. This modest demand has driven significant investment in new production capacity, with 12,000 kg/year recently added by Rio Tinto in Canada, and the Elk Creek project in Nebraska [4] is on the horizon, which has the potential to produce 100,000 kg/year of Sc₂O₃.

You might ask what is driving this demand for scandium? Mainly, it is its use in solid oxide fuel cells, which now represents the majority of the global demand, and which is growing year on year. Its use in high-strength aluminium alloys comes a distant second in terms of volume demand, partly because so little scandium (less than 1%) is needed in each kilogram of aluminium alloy to provide the significant effect that it has on the strength.

Why does scandium have such a beneficial effect on aluminium alloys?

Some aluminium alloys derive strength from precipitation hardening, meaning that they grow small crystals in a matrix of aluminium. Aluminium alloys were regarded as some of the first nanomaterials because the precipitates can be nano-size in scale and act as reinforcements distributed throughout the aluminium matrix. In the case of scandium, it forms Al₃Sc and these precipitates are very fine and, therefore, more evenly distributed throughout the metal, thus assisting with weldability. Technically, this means that the precipitates don’t just hang out at the grain boundaries, which benefits the strength and weldability.

In practice, one cannot get more than 1% of scandium in solution with aluminium, but that small addition has a lot of impact. A 50-100 MPa benefit can be achieved per 0.1 wt.% of scandium added [5]. Not only does it improve strength, but it also benefits corrosion resistance, temperature resistance, stiffness, and formability, like welding – or Additive Manufacturing.

Scandium is a force multiplier of sorts which is good for aluminium. Scandium and zirconium team up when they are together to drive more benefits. Zirconium can be partially substituted for scandium [6], further reducing the demand for scandium.

Who is using it and why?

Aluminium scandium alloys were used early on in aviation components on both MiG-21 and MiG-29. In the 1970s, lightweight bicycle frames were made with Al-Sc materials and were the height of performance until carbon fibre came along. However, those alloys had a relatively small amount of scandium (0.1%-0.3%), so the strength achieved was modest. This was because there was a limit to how much scandium could be kept in solution during the solidification of the melt, and they were at that limit.

Everything changed with the emergence of Additive Manufacturing. Thanks to the very high cooling rates in these processes, due primarily to the small melt pool, it is possible to keep a much higher percentage of scandium in solution and achieve significantly higher strengths. Scalmalloy was the first alloy to exploit this effect and it enabled a combination of strength and ductility which gives even the highest-strength forged aluminium alloys a run for their money.

Combining such a material with the design freedom and low-volume economics of Additive Manufacturing was first exploited in motorsport, initially by Formula One teams, but later by many different race series, making good use of those properties in a wide range of motorsport applications.

Since then, it has been picked up in competitive cycling, with several record-breaking bicycles produced from Scalmalloy. The ability to produce bicycles with the aerodynamics of a carbon fibre bicycle, but tailored to the rider (critical for overall aerodynamics and performance) while achieving a mass very close to that of a carbon fibre bike seems to be a winning combination. Could a ‘Scandium Renaissance’ be coming to the next Olympics?

Scalmalloy has also been used to make very lightweight structures for satellites and systems components for aircraft flight tests, but it isn’t yet at the point of high-volume series production in aerospace, which we all know to be traditionally a conservative field of application. We are pretty confident that this will come next. The ability to substitute so many other alloys in terms of material properties is a massive benefit for aerospace applications and minimises the cost of qualification to cover a wide range of applications of AM.

How much scandium do we need to make a real difference?

As noted previously, very small percentages of scandium are needed – less than 1% by weight. This translates into a relatively small mass required to make Al-Sc alloys. How much scandium is needed to make a big difference to the industry?

In Fig. 3, we imagine a high rate of aircraft production, which delivers about 700 aircraft per year. Now let’s imagine we are processing 100 kg of Scalmalloy parts per aircraft: we would only need about 100 t/yr of Scalmalloy to do that, which translates to only about 1,000 kg of scandium demand per year, which is a tiny fraction of the market. That little bit of scandium can have a big impact, not least of which is the potential for weight saving, which can drive improved efficiency in aircraft and reduce emissions from aviation. Our point is that the scarcity of supply of scandium is not a justifiable concern.

Is it sustainable?

According to the International Aluminium Institute, aluminium is one of the most recycled materials on earth. This is a pretty bold statement, but they cite the global recycling efficiency rate as 75% [7]. Scandium containing aluminium alloys go further because such a small addition of scandium improves material properties dramatically, which reduces the total consumption of other materials as a result. In addition, they are no less recyclable than other aluminium alloys and can be melted again. Through new, sustainable technologies, such as the DirectPowder process developed by Metal Powder Works, AlSc bar can be converted to AlSc AM sized powder directly with an efficiency of approximately 95% which emits far less CO₂ emissions than conventional atomisation.

For discussion…

It is no secret that scandium benefits aluminium alloy performance in strength, temperature, corrosion, stiffness, and weldability. Small amounts of scandium go a long way. Industry is working on solutions to solve current challenges when using cast and wrought aluminium alloys in Additive Manufacturing, but two issues are keeping the market from expanding:

The perceived supply chain risk of scandium

We’ve shown that the supply is sufficient to meet increased demand. Newer methods to recover scandium from tailings are more than sufficient to address the market need. Production via tailings opens up entirely new supply chains in Canada, the United States, Australia, and the Philippines amongst others, thus further minimising supply chain risks.

Part re-design

Part re-design is required when the additively manufactured materials show negative margins. When the material meets or exceeds the strength/stiffness requirements, it is more of a ‘drop in’ replacement, and a re-design is only required to optimise – meaning the new material can simply be dropped in and we’re ‘good to go.’

Other considerations

The reality is that scandium containing aluminium alloys are still in their infancy and much of the research and development that has been done since the 1970s has never been commercialised. Even the widely known AM alloy, Scalmalloy, is only the tip of the iceberg. APWORKS is currently developing multiple variants including Scalmalloy HX (for high-temperature applications), Scalmalloy CX (for cryogenic applications), and Scalmalloy EX (for electrical and thermal applications). And they are not alone, with a number of universities around the world continuing to explore the alloys that can be created using scandium. The potential of these alloys is only growing as more bright minds are working on them.

AlSi10Mg alone is not going to drive major growth in AM. It is a very useful alloy for lower strength applications, but it is not going to enable truly disruptive high strength applications or replace all of the legacy cast aluminium alloys. Fig. 4 shows a comparison of AlSi10Mg alongside the high-strength AM aluminium alloys.

The price of powders is not going to drive volume in AM because it cannot possibly compensate for the other 90% of the cost which is driven by processing. For AM to move to volume production it needs to add value through performance, and it will not do that with ‘average’ materials which merely mimic their conventional counterparts.

There are more alloys for AM being developed, and the best of those will enable the applications that will drive growth in the AM market. Some of those will contain scandium, others will explore other opportunities that AM processing enables. Materials will be the key enabler for volume applications.

Lastly, the atomisation of metals brings about its challenges – primarily economic. The yield of powders for sizes traditionally desired for AM can be as low as 40%. This low yield adds further cost to already expensive aluminium alloy powders. There is hope as the industry is already looking into capturing the oversized powder that has no market and converting it into powder via solid-state means such that the combined cycle drives efficiency. This new efficiency also addresses access to scandium. When the overall yield can grow to 80% from 40%, it will reduce cost.

Conclusions

Requirements are a combination of technical and economic factors. Given the tight relationship between structural performance, design, materials, and manufacture in AM, the right material has to be matched to the right design and utilise the AM machine for as little time as possible. This is even more true in aluminium as the powder is more expensive than mill product, and AM is slower than machining.

It is time to embrace the future with AM and scandium. AM enables using ‘premium’ materials and promotes lighter designs. Using less raw material and enabling lighter designs are inherently sustainable. The industry needs a high-strength aluminium alloy to improve the business case of using AM and aluminium.

We have previously shown our additively manufactured brake caliper is lighter and cheaper in Al-Sc than AlSi10Mg. We have made the case that scarcity or supply chain issues are increasingly being resolved and, in any case, very little scandium is required to begin making an impact. Let’s make the Nilson team proud and make scandium mainstream.

Authors

Jonathan Meyer
CEO
APWORKS GmbH
www.apworks.de

John E. Barnes
Metal Powder Works & The Barnes Global Advisors
www.metalpowderworks.com
www.barnesglobaladvisors.com

References

[1] J Meyer, J Barnes, Scalmalloy is too expensive and design optimisation only makes sense in aerospace. True or false? Metal AM Vol. 5 No. 1, Spring 2019 Available to read in full here: https://bit.ly/4a8VR46

[2] https://en.wikipedia.org/wiki/Scandium

[3] https://en.wikipedia.org/wiki/Abundance_of_elements_in_Earth%27s_crust

[4] www.energy.gov/sites/default/files/2022-08/4.%20DOE_Golden_Conference_NioCorp_Presentation_July-2022.pdf

[5] Chapter 12, Aluminium scandium Alloys, Thomas Dorin, Mahendra Ramajayam, Alireza Vahid, Timothy Langan, Chapter 12, Aluminium scandium Alloys, Editor(s): Roger N. Lumley, In Woodhead Publishing Series in Metals and Surface Engineering, Fundamentals of aluminium Metallurgy, Woodhead Publishing, 2018,

[6] Scandium-enriched Nanoprecipitates in aluminium Providing Enhanced Coarsening and Creep Resistance, Anthony De Luca, David C. Dunand, and David N. Seidman

[7] www.international-aluminium.org/resource/aluminium-recycling-fact-sheet/

[8] Unpublished results, The Barnes Global Advisors, Designing for Additive Manufacturing course.