Researchers from the University of Central Florida, including Abhijeet J. Kale, Sanjeev S. Navaratna, Pratik Sahu, Henry Chan, B. R. K. Nanda, and Rohit Batra, have made significant strides in the field of spintronics, a technology that leverages the spin of electrons to process information. Their work, published in the journal Nature Communications, focuses on accelerating the discovery of two-dimensional (2D) materials with high spin Hall conductivity (SHC), a critical property for efficient spin-charge interconversion in next-generation devices.
The team employed an active learning framework, a type of machine learning (ML) technique, to navigate the vast chemical space of 2D materials. Traditional methods for discovering high-SHC materials are often expensive and time-consuming, but the researchers found a way to streamline the process. They started with a diverse set of 24 2D systems and expanded the dataset to 41 cases over three active learning loops. This approach allowed them to identify several high-SHC candidates, with the best candidate exhibiting a SHC of 271.52 (hbar/e) Ohm^-1, nearly 23 times higher than the top performer in the initial round.
The researchers also uncovered several features that play a critical role in shaping the spin Hall response in 2D systems. These include orbital symmetry near the Fermi energy, types of atomic species, material composition, covalent radii, and electronegativity of constituent atoms. By making their data publicly available, the team aims to facilitate further advances in 2D spintronics, potentially leading to more efficient and powerful electronic devices.
For the energy industry, this research could pave the way for more efficient power conversion and processing technologies. Spintronics has the potential to revolutionize the way we handle information, and the discovery of high-SHC materials could lead to more energy-efficient devices. This could have significant implications for data centers, which consume vast amounts of energy, as well as for the development of new types of sensors and memory devices. The practical applications of this research are still in the early stages, but the potential for improving energy efficiency and performance in various electronic devices is substantial.
This article is based on research available at arXiv.