In the lab, voltage slides graphene layers to switch magnetism

Graphite and diamond are made of the same carbon atoms, yet one is soft and dark and the other is hard and clear. The difference comes from how the atoms are arranged, and changing that arrangement in a material usually means breaking strong chemical bonds and rebuilding them, a slow, energy-hungry process that calls for high pressure and high temperature.

At Cornell, the lab of Kenji Yasuda, assistant professor in the School of Applied and Engineering Physics, tried a different route. The team, led on the experiments by doctoral student Daniel Brandon, worked on a stack of four layers of graphene, a material one atom thick made of carbon atoms in a hexagonal pattern. Their study was published on October 14 in Nature Nanotechnology.

The trick lies in the bonds between the layers. These are van der Waals bonds, much weaker than the bonds inside each layer. Applying a voltage made the layers slide over one another, a movement the researchers call sliding. The stack moved from the most common and most stable arrangement, known as Bernal, which is nonmagnetic, to a rarer and less stable one, called rhombohedral, a state associated with magnetism, superconductivity and other exotic quantum phenomena. Yasuda describes the aim as finding a way to change a material's structure that is more efficient and still useful for technology.

The researchers highlight one feature: the change is nonvolatile. Once the material is in the new state, it stays there, which is why Brandon describes the sliding as a form of memory. The team says the change can be reversed with voltage and that the material goes back and forth between the two states fairly consistently. The study does not report how many cycles the switching survives, nor how it would behave outside the laboratory.

For materials researchers, the result offers a lever that works electrically rather than through extreme pressure or heat, and Yasuda says it could extend beyond magnetism to superconductivity and other quantum states. Applications such as nonvolatile memory remain a prospect that the authors name, not a product. The study gives no timeline and no device, and the demonstration stays, for now, a four-layer graphene stack in a laboratory.