AI helped scientists discover two new superconductors, bringing them closer to a room-temperature breakthrough that could change electronics

Image Credit: Esa Kapila, via Aalto University’s LinkedIn

Artificial intelligence is now fast becoming an important component in scientific investigations, and its recent achievement may have very significant effects on the advancement of technology. Scientists devised a way to utilise artificial intelligence in scientific research, which has not only resulted in the discovery of two new superconductors, but has also expedited the process of discovering even more superconductors. This achievement may bring us one step closer to one of the main objectives of physicists: a room-temperature superconductor. The research led by scientists at Aalto University and their international collaborators, combines machine learning with quantum physics to identify promising superconducting materials far more efficiently than traditional methods. Using this approach, the team successfully predicted and experimentally confirmed two new superconductors, named “YRu₃B₂” and “LuRu₃B₂” proving that artificial intelligence can significantly speed up materials discovery.

What are superconductors and why are they important

Superconductivity refers to a material that can carry electric current with zero electrical resistance below a critical temperature. In contrast to normal conducting wires that lose some energy as heat, no energy will be lost when carrying electricity through a superconducting material.This special ability is applicable for various purposes. Superconductors are used in MRI scanners, particle accelerators and some maglev (magnetic levitating) trains. But most superconducting materials function at very low temperatures approaching absolute zero (-273.15°C). For decades now, researchers have sought the elusive ‘room-temperature superconductor’ – a substance that can achieve superconductivity at normal temperatures without freezing or immense pressure. This would revolutionise fields such as energy transmission, quantum computing, medicine and transport among many others.

How AI changed the search

Conventionally, finding any new superconductor is associated with long experiments and quantum-mechanical calculations that take many years to perform. Many options of combinations of chemical elements have to be tested until finally something proves successful.The proposed study utilises the novel technique of machine learning combined with theoretical physics. Rather than checking each candidate in turn, the machine-learning model goes through huge amounts of data of known substances and identifies the most promising candidates.Researchers then validate the most promising candidates experimentally. Once the AI identified materials with a high likelihood of exhibiting superconductivity; the predictions were first confirmed through theoretical calculations. Collaborators at Rice University, led by Professor Emilia Morosan, then synthesised the materials by chemically combining their constituent elements into entirely new compounds. Laboratory tests confirmed that both YRu₃B₂ and LuRu₃B₂ are indeed superconductors providing strong proof that the AI-guided discovery process works in practice rather than just in theory.The proof-of-concept study describing the discovery was recently published in the journal Physical Review Research.

Image Credit: Image Credit: Esa Kapila, via Aalto University’s LinkedIn

Why this matters for room-temperature superconductivity

Although the newly discovered materials still require cooling and are not room-temperature superconductors; researchers say the real breakthrough lies in the discovery process itself rather than the materials alone. Instead of spending years searching manually, scientists can now use AI to screen millions, or potentially billions, of candidate materials in a fraction of the time. This dramatically increases the chances of identifying compounds that may eventually function at much higher temperatures. The research is part of the SuperC consortium, an international initiative launched in 2023 by Professor Päivi Törmä and a team of leading physicists from around the world. The collaboration brings together experts in quantum physics, materials science and artificial intelligence with the shared aim of discovering new superconducting materials that could help reduce energy consumption and support climate goals. One of the consortium’s long-term ambitions is to identify a practical room-temperature superconductor by 2033 – a breakthrough that could transform the future of energy and electronics.According to Aalto University Professor Päivi Törmä, who leads the SuperC consortium, the approach could largely speed the discovery of new superconductors. The international SuperC consortium, which is leading much of this research, hopes that AI-assisted materials discovery will help achieve the long-standing goal of finding a practical room-temperature superconductor within the next decade. While that target remains ambitious, the successful discovery of two new materials demonstrates that the strategy is already producing real results.Explaining the significance of the research, Professor Päivi Törmä said, “Superconductive materials that can operate at room temperature would forever change the way we consume energy. If such a material could replace regular conductors in applications like computers and data centres, global energy consumption could be slashed and the heat footprint of the ICT sector vastly reduced.”

Could this change everyday technology

If scientists discover a room-temperature superconductor, it would mean a revolution in many ways. Power grids would be able to transfer electricity almost without any losses increasing their efficiency greatly. Electric cars would become lighter and more efficient, while computers would work much faster and consume much less energy.Quantum computers that nowadays require complex cooling systems would become more applicable. Medical equipment such as MRI scanners would be easier and cheaper to manufacture and use, and magnetic levitation systems could be made for ordinary usage.Although the mentioned advancements seem far from realisation, the newest discovery can be considered a breakthrough. Not only lab tests are conducted by people anymore, scientists have started using artificial intelligence as a research partner that is able to analyse and find new materials much faster than before.Therefore, the discovery of YRu₃B₂ and LuRu₃B₂ is important not only because they are the new superconductors, but also because it opens a completely new perspective for scientific researches. As AI technology develops further, it can potentially help researchers solve some very complicated problems in the materials’ research field.