Giant Spin-to-Charge Conversion by Tailoring Magnetically Proximitized Topological Dirac Semimetal

In the quest to build faster, more energy-efficient computers, scientists are looking for ways to control magnetic information using electric currents rather than magnetic fields. This approach, known as spintronics, relies on a property of electrons called "spin," which can be thought of as a tiny internal magnet pointing either up or down. A major challenge in this field is converting the flow of these spinning electrons into an electric current, and vice versa, with high efficiency. For years, researchers have turned to a special class of materials known as topological insulators and semimetals. These materials possess unique surface states where electrons are locked in a specific relationship between their direction of travel and their spin, acting like a highly efficient highway for information. However, to use these materials in real devices, they must be placed next to a magnet. The problem is that the magnetic influence from the neighboring magnet often disrupts the delicate surface states, effectively closing the highway and ruining the efficiency.

A team of researchers at the University of Tokyo and other institutions has now found a way to navigate this problem without using extra layers to shield the materials. They studied a specific combination of a magnetic metal, iron, and a topological semimetal, a form of tin known as alpha-tin. Instead of trying to block the magnetic influence, they investigated how the thickness of the tin layer changes the behavior of the system. By growing layers of tin ranging from very thin to nearly forty nanometers thick and placing a thin film of iron on top, they discovered a surprising sweet spot. When the tin layer reached a thickness of twenty-five nanometers, the system exhibited an exceptionally strong ability to convert spin into charge. This peak performance occurred because the magnetic influence from the iron layer suppressed the topological states on the top surface of the tin but left the states on the bottom surface intact and functional.

The researchers measured this effect using a technique called spin pumping, where they shook the magnetic layer with radio waves to inject a stream of spinning electrons into the tin. As these electrons traveled through the tin, they generated a voltage that could be measured. The team observed that the strength of this voltage signal did not simply grow with the thickness of the tin; instead, it rose to a maximum at twenty-five nanometers and then fell off as the layer got thicker. This behavior provided a crucial clue. If the conversion were happening at the top surface where the iron sits, the signal would have been weak or non-existent due to the magnetic disruption. If it were happening in the bulk of the tin, the signal would have grown steadily. The fact that it peaked and then dropped suggested that the conversion was happening at the bottom interface, between the tin and the substrate it was grown on, but that the electrons had to travel through the tin to get there.

To understand why this happened, the team combined their experimental data with computer simulations of the material's electronic structure. The simulations revealed that the magnetic influence from the iron layer, known as the magnetic proximity effect, penetrates deep into the tin. At the top interface, this effect destroys the special topological states. However, at the bottom interface, far enough away from the iron, the states remain, though they acquire a small energy gap due to the long-range magnetic influence. The researchers found that at twenty-five nanometers, the tin layer is thick enough to allow the spin current to reach the bottom surface effectively, but not so thick that the signal dies out from traveling through the material. This specific thickness allowed the bottom surface states to act as a highly efficient converter, generating a voltage that was among the highest ever recorded at room temperature for any material.

The study explicitly ruled out several alternative explanations for this high efficiency. The researchers confirmed that the signal did not come from the iron layer itself, nor from the top surface of the tin where the iron sits, because samples with very thin tin layers showed no such conversion. They also determined that the signal was not a result of heat or other parasitic effects, as the voltage increased linearly with the power of the radio waves only in the sample with the optimal thickness. Furthermore, they showed that the bottom surface states were the dominant source of the conversion, a conclusion supported by the negative sign of the voltage, which indicated the opposite spin direction compared to previous studies that used a spacer layer to protect the top surface.

By tailoring the thickness of the tin layer, the researchers demonstrated that it is possible to harness the power of topological materials even when they are in direct contact with a magnet. They calculated that the efficiency of this spin-to-charge conversion, measured as an inverse Edelstein length, reached a value of negative 3.1 nanometers. This is a remarkably high number, comparable to the best results seen in other complex material systems, but achieved here without any insulating spacer layers. The work suggests that the magnetic proximity effect, often seen as a nuisance, can be managed by engineering the geometry of the material to preserve the useful states on the far side of the layer. This finding offers a new path for designing spintronic devices that are simpler to build and potentially more efficient, as they do not require the extra steps of inserting protective barriers between magnetic and topological components.