In a field that sometimes rediscovers its own past, a Swedish materials physicist and her colleagues have turned to artificial intelligence to dig through the forgotten archives of earlier chemistry. Among the records they searched, they found dozens of two-dimensional materials called MXenes that had already been made but never recognized, some reportedly described decades ago. Their discovery lends weight to an old lesson: much of what we seek may already be sitting quietly in the libraries we have stopped reading. Whether these thin, many-layered materials will ultimately serve water, energ
AI Uncovers 33 Forgotten MXene Materials in Old Databases, Plus 11 New Ones
The literature and databases are treasure troves
Mark
Start with the odd part. How do you discover something that already exists?
Mimi
The team used AI to search older materials databases and found 33 MXenes that had been made before but that the field wasn't aware of. Rosén says researchers sometimes claim discoveries that were made decades earlier, in Germany in the 1970s, for example.
Luke
Worth noting that the German example is her illustration of the general problem. The reporting doesn't say the 33 materials all trace to that era or place.
Mark
Fair. And the 11 new ones?
Mimi
The team made those after the database findings. They're new compositions, which is where Shamshirgar sees more room to explore uses.
Mark
And the manufacturing method?
Mimi
No external heating, a few seconds, fewer hazardous chemicals, less water. That's the part Rosén ties to scaling up.
Luke
Scaling is her expectation, though. The source says the research front is moving, which suggests scale-up is possible, not that it has been done.
Mark
What about the applications, water purification, energy storage and so on?
Mimi
Those are possibilities from the layered structure and the chemistry between layers. Some uses haven't even been investigated yet.
Luke
So treat them as potential. No performance figures are given, and we don't know how the 33 or the 11 compare on any measure.
Mimi
The rare earth substitution is the concrete addition: it can make MXenes magnetic, which widens the possible uses.
O Pulso
- Dozens of MXenes once thought unknown surfaced from old databases, challenging the field's sense of what it had already discovered.
- Researchers at Linköping University used AI to sift through archived records, uncovering 33 overlooked materials that some reportedly dated back decades.
- A new synthesis route produced 11 fresh MXenes in seconds, with no external heating and far fewer hazardous chemicals and less water than before.
- Scaling up the fast method is now the decisive hurdle, since potential uses in water purification, energy storage, and biomedicine remain largely untested.
In a field that sometimes rediscovers its own past, a Swedish materials physicist and her colleagues have turned to artificial intelligence to dig through the forgotten archives of earlier chemistry. Among the records they searched, they found dozens of two-dimensional materials called MXenes that had already been made but never recognized, some reportedly described decades ago. Their discovery lends weight to an old lesson: much of what we seek may already be sitting quietly in the libraries we have stopped reading. Whether these thin, many-layered materials will ultimately serve water, energy, or medicine depends on whether they can be made at a scale that matches their promise.
Johanna Rosén, professor of materials physics at Linköping University, speaks bluntly about a recurring embarrassment in her field: researchers sometimes announce breakthroughs that were already accomplished years earlier, such as work done in Germany in the 1970s. Her group decided to go looking for that buried work using artificial intelligence.
Working with colleagues in Estonia and Finland, the team had AI search existing materials databases for MXenes, a family of two-dimensional materials first discovered in 2011, that had been made but never recognized by the field. The search turned up 33 such materials in the older records. Inspired by these finds, the researchers then created 11 new MXenes themselves. The work, published in Science with Ali Saffar Shamshirgar as first author, was informally called the Treasure Chest Project. Rosén's image was simple: the literature and databases are treasure troves, and one only has to open the lid.
The appeal of MXenes lies in their structure. They are only a few atomic layers thick, giving them vast surface area for their weight, and their layers can be stacked so densely that a film a few millimetres thick may contain millions of them. Chemistry between those layers varies with composition, opening possibilities in water purification, energy storage, telecommunications, and biomedicine, as well as uses not yet explored.
The study also introduced a new way of making these materials. The route requires no external heating, takes only seconds, and uses significantly fewer hazardous chemicals and less water than earlier methods. The researchers further showed that rare earth metals can replace the usual metals in MXenes, potentially making them magnetic.
Rosén sees scale as the deciding factor. Many applications remain promises rather than demonstrations, and the open questions are whether this fast method can be enlarged for industry and which of the newly identified compositions prove useful in practice.
Johanna Rosén, Professor of Materials Physics at Linköping University in Sweden, has a blunt way of describing a recurring problem in her field. Researchers, she says, sometimes reinvent the wheel and announce great discoveries, only to find the work was already done, in Germany in the 1970s, for instance. Her group has now used artificial intelligence to go looking for exactly that kind of buried work.
Working with colleagues in Estonia and Finland, the team had AI search existing materials databases for MXenes that had already been made but that the research field did not know about. MXenes are a family of 2D materials first discovered in 2011. In their basic form they combine a metal with carbon or nitrogen atoms. What matters most is their structure: they consist of only a few atomic layers, which gives them an extremely large surface area relative to their weight.
The search identified 33 MXenes that already existed in the older records. Prompted by what they found, the researchers then made 11 new ones themselves. The work is published in Science, and Ali Saffar Shamshirgar, a researcher at LiU, is first author of the article. The team informally called the effort the Treasure Chest Project. In Rosén's words, the literature and databases are treasure troves, and you just have to open the lid and see what is inside.
The layered structure is central to why the materials attract interest. MXenes can be stacked, and a film only a few millimetres thick may hold several million layers. Between those layers a wide range of chemical reactions can take place, depending on the composition of the MXene. That gives the materials properties that may suit water purification, energy storage, telecommunications and biomedicine, along with other uses the researchers have not yet investigated. Shamshirgar says MXenes can have a broad range of useful properties, many with potential applications in sustainability, and that new compositions open up more possibilities for how they can be used.
The study's other major result concerns how the materials are made. The team produced MXenes by a completely new route that needs no external heating and takes just a few seconds. Compared with earlier approaches, it requires significantly fewer hazardous chemicals and less water. Rosén says synthesis is an active area of research, which suggests manufacturing can be scaled up.
She regards scale as the deciding factor. In her view, scaling up is what will push MXenes over the edge into significantly wider use in applications.
The researchers also showed that rare earth metals can replace the metals normally found in MXenes. That substitution can make the materials magnetic, which would broaden the range of possible applications further.
Much of the application list remains potential rather than demonstrated. The next questions are whether the fast synthesis method can be scaled, and which of the newly identified and newly made compositions prove useful in water treatment, energy storage or medicine.
Citações Notáveis
I think it's the scaling up that will push MXenes over the edge to actually being significantly more used in different applications.— Johanna Rosén, Professor of Materials Physics, Linköping University
Discovering new compositions gives us the opportunity to explore even more possibilities for how these materials can be used.— Ali Saffar Shamshirgar, researcher at LiU and first author
Quer a matéria completa?
Leia o original em chemeurope.com ↗
Science & Space