Scientists in Australia have developed the largest quantum simulator to date, using 15,000 qubits to model exotic quantum materials. This device, known as Quantum Twins, could help optimize superconductors and other advanced substances. Built by embedding phosphorus atoms in silicon chips, it offers unprecedented control over electron properties.
Michelle Simmons and her team at Silicon Quantum Computing in Australia have unveiled Quantum Twins, a quantum simulator composed of 15,000 qubits arranged in a square grid. This marks the biggest such device yet, surpassing previous arrays made from thousands of extremely cold atoms. By embedding phosphorus atoms into silicon chips, the researchers transformed each atom into a qubit, allowing precise arrangements that mimic atomic structures in real materials.
The simulator enables detailed control of electron properties, such as the difficulty of adding an electron to a grid point or enabling electron 'hopping' between points. This capability is essential for understanding electricity flow in materials. Simmons noted, “The scale and controllability we have achieved with these simulators means we are now poised to tackle some very interesting problems.” She added, “We are designing new materials in previously unthought-of ways by literally building their analogues atom by atom.”
In tests, the team simulated a transition between metallic and insulating behaviors in a mathematical model of how impurities affect electric currents. They also measured the system's Hall coefficient across temperatures, revealing responses to magnetic fields. Conventional computers struggle with large two-dimensional systems and complex electron interactions, but Quantum Twins shows promise here.
Looking ahead, the device could explore unconventional superconductors, which operate under milder conditions than traditional ones but require deeper microscopic insights for room-temperature applications. It may also investigate metal-molecule interfaces relevant to drug development and artificial photosynthesis. The findings appear in Nature (DOI: 10.1038/s41586-025-10053-7).
