October 2024

In a recent sturdy, researchers at the U.S. Department of Energy’s (DOE) Argonne National Laboratory and Northern Illinois University discovered that they could use light to detect the spin state in a class of materials called perovskites (specifically in this research methylammonium lead iodide, or MAPbI₃). 

To understand spin, consider electrons orbiting the atomic nucleus. When atoms are close together, they can share some of their outer electrons, which creates a bond between them. Each bond contains two electrons that are ​“paired,” meaning they share an orbital — the region where they move. Now, each of these paired electrons has one of two possible spin states: spin up or spin down. If one electron is spin up, the other is spin down. Since we can’t know exactly which electron has which spin without looking at them, we say they exist in a quantum superposition — a state where they are both spin up and spin down until observed.

TDK Corporation has announced the development of a neuromorphic element called a “spin-memristor” that has very low power consumption. By mimicking the energy-efficient operation of the human brain, this element could cut the power consumption of AI applications down to 1/100th of traditional devices. Collaborating with the French research organization CEA (Alternative Energies and Atomic Energy Commission), TDK has shown that its “spin-memristor” can serve as the basic element of a neuromorphic device. 

Going forward, TDK will collaborate with the Center for Innovative Integrated Electronic Systems at Tohoku University on the practical development of the technology.