The paper: https://www.science.org/doi/10.1126/science.aec4995
As an aside, “super alloy” is not the best wording choice on the part of the author of this sciencealert article, superalloys are an established alloy family that follow a different design strategy and have a very different composition profile https://en.wikipedia.org/wiki/Superalloy
This is sleight-of-hand.
For metals, the operative properties are usually ultimate tensile strength and tensile yield strength. Compressive strength is typically a non-factor for most engineering alloys; only concrete is judged by its compressive strength, and sometimes various engineering ceramics. (e.g. SiC, compressive strength = 3.9 GPa.)
About ten years ago, there were a lot of papers on amorphous metal alloys that had "extreme strength" -- compressive strengths in the 5-6 GPa range -- but tensile strength was not reported and very low. Some measure of ductility was also present, but it too was very low. Those amorphous alloys were classical brittle materials; more ceramic-like than metal-like. I fear the same is probably the case here, with the alloy potentially fracturing along crystal type grain boundaries.
Until they report actual tensile strength and elongation, don't believe the hype. High compressive strengths are not very useful.
Statements like this had me confused for years about how steel is made. Steel is raw iron ore with some carbon removed. (Among other things) Pure iron isn't common in nature, it's usually found instead with too much carbon. That is, to have the properties we like in steel.
Ugh. The most basic bitch metallurgy discussion possible.
Oh! It’s stronger than aluminum?! So is bronze, we’ve hard that for awhile! Is the new material lighter than aluminum while being stronger? Is it corrosion resistant? Is it machinable? Can you weld it? Does it oxidize? Does it lose all its strength under moderate heat? Does it temper, do you have to temper it? Is it inert? Can it extrude? Can it be formed into billet or just plate/bar? Does it shatter?
Oh but 2x stronger than <some steel> and 3x stronger than <some aluminum>… is that 2024 aluminum? 6061 common, 7075 aero? Is the steel cold roll or 600-series inconel?
This is an area where if you don’t know what you are talking about, STFU, because anything you say is just going to be embarrassing. This is a you don’t know what you don’t know topic.
As to “high entropy metals”, I’ve heard about this for awhile, I would expect it to be stupid low yield, stupid expensive, and hard to use. There is probably some grade-40 titanium ultra alloy that could make the same “strength” claims but no articles about it because it’s “cost prohibitive”.
… I count this as clickbait metallurgy. No thanks.
> It's two times stronger than steel, three times stronger than aluminum, and twice as strong as the same alloy made in a conventional way.
The source paper in Science, fwiw:
https://www.science.org/doi/10.1126/science.aec4995
And as a personal exercise in intellectual humility, I cast my eyes over the supplementary materials (as those are free-to-the-public)… I’d recommend it:
https://www.science.org/doi/suppl/10.1126/science.aec4995/su...
I get a huge thrill out of looking at serious work outside my expertise. When I’m tempted to imagine the proposition is as simple as it seems from the headline (or the article, or the editor’s note, or the abstract), it excites me to remember just how deeply and carefully and thoroughly people think through things I barely understand.
I mean, it depends on the use case right? Modern tall buildings/skyscrapers with metal framing use metal pillars where high compressive strength is very useful.
Instead of one invincible giant robot, we will have invincible huge swarms of dwarf robots, which will be much more dangerous than any single entity.
"Steel" is defined as carbon content between .02% and 2%. If the metal has less than that it's wrought iron. If it had more than that it's cast iron.
So steel is a range of carbon contents that gives particularly useful properties.
Historically the lower end of that was higher but with that addition of other alloys we have "low carbon steels" that behave like steel with very little carbon.
There is only one new fact in this research, which is however very interesting.
By mechanical working of the alloy, its crystal structure has changed to a mixture of grains with 3 different crystal structures, instead of being composed of homogeneous grains, like most solid-solution metallic alloys.
This change of crystal structure induced by mechanical working has doubled the strength of an already strong alloy, which is the novel and interesting result of this research.
The reason why this could happen is that the alloy contains similar amounts of metals that normally have distinct crystal structures. In alloys with a dominant metal, the alloy tends to have the same crystal structure as that metal, but in this alloy none of the crystal structures is preferred, so the grains take one of them randomly, during deep mechanical deformation.
For compressive strength alone to matter, you’d essentially need a solid steel column - which is so prohibitively heavy, it would often end up collapsing under it’s own weight (in tensile failure, most likely) before it got up to a useful height.
It's hard to know just how much stronger this new processing of the alloy is than other common high strength alloys, as they list compressive yield and not tensile yield strength ... that's if the person writing didn't get the two terms confused.
As a note, I use duckduckgo and smirked somewhat at its search assist results for the few efforts to find the compressive yield of Bisalloy 400 (something I've had to drill) - checking out the listed sources it was clear it had mistakenly used the tensile yield ...
As an illustration for the differences, I found a page [2] for 4140 alloy and similar yield strengths. 4140 is reasonably workable, drilling isn't the greatest amount of effort either before it's tempered and annealed.
[1] https://www.practicalmachinist.com/forum/threads/milling-mp3...
[2] https://amesweb.info/Materials/Steel-Tensile-Yield-Strength-...
There’s some precedent for mechanical deformation to get nanocrystalline grain structure, and some precedent for mechanically induced phase transformations (see TrIP steels) but I consider both of those concepts pretty advanced
TrIP is probably the closest thing, but I’m not sure how widely known there are among the “metal forging” community? TrIP is usually targeting well known phase transformations to two-phase microstructures, here we have three nanostructured phases.
Finally high entropy alloys are absolutely not well understood, even if the idea of mixing a lot of elements and getting a disordered solution seems simple on its face.
I am no metallurgist, or machinist, or work in fabrication, at best I can qualify as a welder made by grinder and paint. Yet, reading that feels more like a sci-fi story (and even technologies that haven't been imagined yet) than just sci.
And yes, when we design beams, we check the stresses in both tension and compression. For a symmetrical shape like an I-shaped beam, both verification will be the exact same formula so we don't need to calculate both explicitly.
For columns, we do check the compressive stress, and it is the controlling failure mode for short columns. Long, slender columns will buckle before the compressive stress exceeds the allowable limit.
Yes, I can force myself to act unnaturally to an extent. I even did. What I'm saying is that I wish I didn't need to, because it sucks.
Is that correct? Pardon for asking a bit of a question to your comment. I don't know too much about growing crystals for things like jet engine blades.
They are only the same in an abstract theoretical sense, not an actual one.
And bucking within the column (and/or the resistance too it) is primarily resisted by tensile strength.
Also, the price of a Ta-Hf-Zr-Nb-Ti alloy isn't especially elastic. Tantalum in particular is an unavoidably expensive element.
I'll grant that what would be exceptional is a metal with a >2GPa compressive strength, a >2GPa tensile strength, and decent damage tolerance and ductility. I don't think that this paper describes a material with that outstanding combination of properties, though -- it's much more likely to describe a simple brittle material.
What you said was like 99% completely false.
>that compressive stress limits are much lower in steel than tensile
No they are not.
> as you’re noting
I never acknowledge that. Your lack of understanding in what column buckling is doesn't change anything. Unironically, you should ask chatgpt, it can probably give you a high level explanation that is more compatible with your understanding of physics than what I can give you.
Buckling is a stability problem, not a resistance problem. The compressive stress resistance is literally not part of the formulas that we use to verify it. The column could have 100 MPa, 350 MPA, or 359918 MPa compressive resistance and it would change nothing when it comes to buckling. Only the elastic modulus and the physical shape of the column (length, inertia, etc.) is relevant for buckling verification.
>And bucking within the column (and/or the resistance too it) is primarily resisted by tensile strength.
Than please explain to me why we don't need the tensile (or compressive) resistance of the material to know the buckling resistance of the column. I really want to hear that one.
In all fairness though, if your enemies are just Afghan tribemen of Vietnamese farmers with an AK you don't have much of an excuse for losing. But even then you can can the moral high ground because you value life more than them.
Yes, it's all propaganda. But they are distinct tools (approaches?). It's fine to say your enemy has better equipment though, because that gives you an excuse for more R&D and gives you an excuse if you didn't win, why the win took longer, why the win was 'honourable' despite the win seeming inevitable by every conventional metric.
To be slightly more kind. I suppose you could argue this helps the peace afterwards.
If you say the beaten adversary was a worthy opponent, they can help keep their heads high. The Allies are now friends with Germany and Japan. I'm sure some reframing comes along with that.
Not sure if troll or misguided, but you know there are people still alive today that can talk about that moral superiority from experience, right? Just ask them, read their stories, etc. If this is how we're talking about that part of the world in that time-frame, I shudder to think how it'll be talked about when there aren't any more people from that period around. We kinda can see that with the neo- movements...
The easiest way to visualize Young’s modulus is typically using a stress/strain curve from a tensile force test, but can also be visualized by a compressive force test.
Young’s modulus is the elastic region that is ‘under’ both curves, since it’s a value for how much something can safely bend* and return to it’s original shape without permanent damage, and bending causes both compressive and tensile forces in a material.
The lowest value of the two upper limits - of course - will set the upper elastic limit!
Specifically, to a very simple approximation, the ‘inside’ of a bent beam will be limited by compressive strength, the ‘outside’ by tensile. The overall beams strength will be whatever the lowest of the two values is, as that is when the beam will fail somewhere, hence, that what you want to use in civil engineering, eh?
*at a first approximation. It is of course much more complex than that.
For a column, the limit is the lowest of the ability of the column to prevent deflection (tensile) and bear the load without ‘pancaking’ (compressive strength). Pancaking is resisted by the full cross sectional area of the column, while deflection is generally only resisted by (to a rough approximation - greatly depending on the actual geometry, as I noted in my prior comment!) half the thickness of the column, so tensile strength of the material is usually the limiting factor for most simple columns of non-trivial length, where unsupported column deflection is the dominant failure mode.
Notably, this is why concrete often uses rebar in civil engineering, as concrete has trash tensile strength on it’s own and requires massive volumes to have sufficient tensile strength to form acceptable beams or unsupported columns, or use only very limiting forms to ensure only compressive forces actually occur. Unreinforced Concrete’s Young’s modulus is trash for this reason.
Which for civil engineering is the best bet!
Also, add a good safety factor on top due to all the other materials fundamentals that they apparently don’t teach in civil engineering school?
Which hey, I get, because they’d be too overwhelming eh? And they’re generally drowned in the noise at that scale anyway.
Depending on the sub discipline, mech-e, aero-e, etc. will of course have to know these things at a finer level of detail.
Someone computing fuselage thickness, designing an engine connecting rod or turbine blade, etc. needs to know what is going on at a much finer level eh?
Those folks will also have to quantify the type of failure modes they expect (fatigue limit, tensile failure, compressive failure, wear, etc.) and design around it.
Young’s modulus isn’t fully useless in those usages, but more specific values tend to be far more useful, as Young’s modulus is fundamentally the lowest value of a mix of material properties.
But then, you’d know that if you looked at my profile eh?
And if you actually knew what you were talking about. Care to quantify your ‘99%’ wrong comment?
The microstructural evolution of the alloy, heated for 32 hours (left) versus 64 hours (right). (Zhang et al., Science, 2026)
Metal alloys are used everywhere from aircraft to cutlery, making them an indispensable part of modern life.
Scientists are continuing to try to find ways to improve them – which often comes down to the way they're initially formed.
Steel is one of the classic alloy examples: mostly iron with a dash of carbon and other elements, making it much stronger and harder than iron on its own.
Now, an international team of researchers has come up with a new way of building alloys. The method, described in a new paper published in Science, promises to make metals that are several times stronger than the materials we rely on today.

The researchers prompted ordered atoms in their alloy. (Monash University/AI)
The trick is using lower, more controlled temperatures than is normal for alloy manufacturing, and letting the metal 'bake' for a specific period.
This leads to a more stable and ordered configuration of atoms, set in blocks known as grains, that are both smaller and more well-packed than usual.
"For more than a century, alloy development has focused on composition and processing," says materials scientist Jian-Feng Nie from Monash University in Australia.
"Our work suggests that how atoms organize during manufacturing may be just as important.
"The real significance is not just this particular alloy, but the demonstration that atoms can self-organize into defect-free structures in a bulk metallic material, meaning a large, continuous piece of metal, not a thin coating, film or microscopic sample."

The alloy was strongest after 32 hours of heating (panel C). (Zhang et al., Science, 2026)
That note on scaling is important – the idea of smaller, better-organized grains has been explored before, but scaling it up into something usable is challenging.
In the new study, the researchers mixed five metals together: hafnium, niobium, tantalum, titanium, and zirconium. After a brief high-temperature melting stage, the alloy was dropped to a relatively low 550 °C (1,022 °F) and left for several hours and even days.
At around 32 hours was when the researchers got their best result: a 'super alloy' called a Refractory High-Entropy Alloy (RHEAD).
It's two times stronger than steel, three times stronger than aluminum, and twice as strong as the same alloy made in a conventional way.
"By carefully controlling how the atoms organize during processing, we were able to create a highly connected structure with exceptional strength and stability," says materials scientist Yu Zhang from Chongqing University in China.
Both the choice of metals and the method of preparation create the conditions for the alloy atoms to organize themselves into repeating grain patterns, responding to the natural stresses between the mixed materials to create a structure free from defects.
That organization, plus the lack of defects and gaps between the recurring grains, is what gives the added strength.
Tests showed the new alloy achieved a compressive yield strength of more than two gigapascals while retaining its ductility, meaning it bends without breaking.
"If this concept can be applied more broadly, it could open the door to materials with properties that were previously considered unattainable, with implications for alloy design that could be applied across many systems and industries," says Nie.
"Instead of increasing alloy content to achieve better performance, we may be able to design internal structures that deliver superior properties with fewer alloying elements. That could lead to more efficient, sustainable, and cost-effective alloy production."
The researchers say their discoveries open up a wealth of possibilities for future manufacturing, in everything from aerospace to energy systems – and even technologies that haven't been imagined yet.
Related: Strange Metal From Beyond Our Planet Spotted in Ancient Treasure Stash
There's a lot more work to do though. Next, the team wants to understand not just what the atoms are doing in terms of rearranging themselves, but why they're doing it, which should enable this new technique to be expanded and refined.
"For more than a century, advances in alloys have come from altering the chemical composition and processing, guided largely by empirical trial and error," says Yiannis Ventikos, the Dean of Engineering at Monash University, who was not directly involved in the study.
"This research suggests we can actually engineer how atoms organize themselves, creating opportunities to develop materials with capabilities that were previously out of reach."
The research has been published in Science.
This article was fact-checked by Clare Watson and edited by Peter Dockrill. While we pride ourselves on our process, we are only human. If you spot a mistake, please let us know.