Out-of-equilibrium excitation can enable efficient spin injection from a ferromagnetic metal (Co) into a semiconductor (MoS2).

© 2019 A*STAR Institute of High Performance Computing

A shot in the arm for spintronics

30 Apr 2019

A spin injection technique that is more than 10,000 times more efficient than existing methods could allow for the development of ultrafast spintronic computers

A simple and efficient method for transferring the currents produced by spinning electrons from metals to semiconductors has been developed by A*STAR researchers, laying the groundwork for faster and more energy efficient spintronic computers.

Although advances in electronics have led to increasingly powerful computers, the electrical resistance of silicon limits the rate — known as the clock speed — at which computers can process information to just a few gigahertz or several operations per nanosecond.

Using the spin of electrons instead of their charge to convey information could overcome this limitation. Additionally, these so-called spin-currents consume less energy than electric currents, making spintronic computers more energy efficient than conventional electronic computers.

Transferring spin-currents from one material to another, however, has proved very challenging as the currents can become scrambled during transportation, destroying the information they carry.

Now, Justin Song from A*STAR’s Institute of High Performance Computing and colleagues from the Institute of Materials Research and Engineering, in collaboration with an international team of researchers, have developed a simple technique that significantly improves the efficiency at which spin-currents are injected from a magnetized metal into a semiconductor.

“While methods exist for spin injection from metals into semiconductors, a key issue is that they typically operate close to equilibrium,” explains Song. “As a result, when a simple ferromagnet/semiconductor interface is used, the spin injection efficiency is low.”

(From left to right) Justin Song, Liang Cheng, Elbert Chia, and Marco Battiato.

© 2019 A*STAR Institute of High Performance Computing 

By heating a ferromagnet, made from cobalt, with pulses of light from a laser, the researchers were able to create spin-polarized electrons with a range of different energy levels in the cobalt, a state referred to as ‘out-of-equilibrium’, and in which their spins mostly point in the same direction.

This out-of-equilibrium state generates spin-current pulses that last less than one trillionth of a second – enough time for them to diffuse from the cobalt into a semiconductor made from single-atom layers of molybdenum disulfide, which then converts the spin-current into an electric current.

“Possibly the most striking aspect is that all this was demonstrated using a simple metal-semiconductor interface, without the complicated and costly structural engineering used in other spintronics,” says Song.

The work establishes a new and simple technique for spin injection that is more than 10,000 times more efficient than what’s available now, and represents a significant step towards the development of ultrafast spintronic computers that use spin-currents as carriers of information, with clock speeds more than a thousand times faster than those in existing conventional computers.

“We are now investigating how pushing systems into an out-of-equilibrium state can lead to effects not possible at equilibrium,” says Song.

The A*STAR-affiliated researchers contributing to this research are from the Institute of High Performance Computing and the Institute of Materials Research and Engineering.

Want to stay up-to-date with A*STAR’s breakthroughs? Follow us on Twitter and LinkedIn!


Liang, C., Xinbo, W., Weifang, Y. Jianwei, Y., Ming, Y. et al. Far out-of-equilibrium spin populations trigger giant spin injection into atomically thin MoS2Nature Physics advance online publication, DOI: 10.1038/s41567-018-0406-3, 21 January 2019. | article

About the Researcher

Justin Song

Senior Scientist

Institute of High Performance Computing
Justin Song received his PhD degree from Harvard University in 2014 and joined A*STAR's Institute of High Performance Computing (IHPC) in 2016 after spending two years at Caltech as a Burke Fellow and Sherman Fairchild Scholar in Physics. His research interests cover a broad range of key themes in theoretical condensed matter physics and emergent phenomena in quantum materials. Some current projects his lab is working on include novel charge/valley/spin/energy transport, topological materials, and 2D layered heterostructures. Song has received a number of awards for his research, including a National Research Foundation fellowship, a Caltech prize fellowship in physics, and an A*STAR National Science Scholarship. He also holds the appointment of Nanyang Assistant Professor at Nanyang Technological University's School of Physical and Mathematical Sciences.

This article was made for A*STAR Research by Nature Research Custom Media, part of Springer Nature