Low-power electronics with a spin
The Graphene Flagship speaks to Paolo Perna, leader of the SOgraphMEM consortium, about graphene's role in next-generation computer memory
Modern-day technology, like smartphones, personal computers and fitness devices, produces a tremendous amount of data that our computers need to process very quickly. This means there is a persistent, rapidly increasing demand for new, fast, low-power devices with an enhanced ability to store information. Graphene-enabled spin logic devices could surpass Moore's law predictions and shape the future of computing.
The Graphene Flagship interviewed Paolo Perna, Group Leader at Graphene Flagship partner IMDEA Nanociencia, in Spain, who leads the SOgraphMEM consortium funded as part of the 2019 FLAG-ERA Joint Transnational Call. SOgraphMEM stands for Spin Orbit functionalized GRAPHene for resistive-magnetic MEMories.
SOgraphMEM also involves Graphene Flagship Associate Members ALBA Synchrotron (Spain), NaMLab GmbH, Jülich Research Centre (Germany), SOLEIL (France) and the Catholic University of Leuven (Belgium), in addition to Graphene Flagship partners CNRS and Thales, France.
Why are spin logic devices so interesting?
In spin logic devices, the basic logic operations are performed by exploiting the spin of electrons. Spins are used as bits – the units of digital information. Ones and zeros are represented with "up" and "down" spins, whereas conventional electronics rely exclusively on the flow of electric charges. The combination of spin and electronics, called spintronics, can reduce the energy consumption of electronic devices, and make them faster.
However, to operate with spins, we need to be able to control and manipulate them. For example, we might want to modify their collective direction, or select a particular spin with a spin filter. One of the most efficient ways to manipulate electron's spins is to exploit the interaction between spin and orbital motion: the spin–orbit (SO) coupling. This leads to interesting new phenomena and underpins another field of electronics, called spin-orbitronics.
What are the applications of spin-orbitronics?
Typical examples are SO-based magnetic random-access memories (SO–MRAMs), which one day could replace our computer memory architectures, such as the RAM, ROM and cache.
While current devices write and read information via mechanical processes with limited reliability and speed, we can exploit SO interactions and engineer materials where the spins align and whirl in peculiar configurations, forming the so-called topological spin textures. Stable against thermal fluctuations, these are only a few nanometers in size, and could become our new-generation bits.
What are the goals of the SOgraphMEM project? And why did you choose to use graphene?
The overarching objective of SOgraphMEM is to design, fabricate and test electrically tunable SO devices, combining the properties of graphene, magnetic materials and ferroelectricity.
We aim to choose materials that can create stable topologically spin textures with long spin lifetime and propagation. We demonstrated that graphene, coupled with ferromagnets and heavy metals, meets all these requirements, even at room temperature.
In addition, graphene allows for the efficient tuning of the device's magnetic properties by electric fields.
Can you tell us about your most recent accomplishments?
We created devices with a thin sheet of a ferromagnetic metal, cobalt, sandwiched between layers of heavy metal (HM) and graphene. The spins at the interface between graphene/Co and Co/HM were stabilised by local exchange interactions, known as the Dzyaloshinskii–Moriya interaction, creating a chiral and stable spin texture in the device.1
By adding a ferroelectric layer – a material that retains the electric polarization – on top of graphene, we will be able to modify the magnetic properties of the system with an external electric voltage: a step towards resistive switching memories.
What's the next step for this research?
In the very near future, we will use our graphene-based devices for spin-caloritronics. We want to modify the spin-dependent properties with thermal gradients. We intend to study voltage-controlled neuromorphic devices, which mimic the way the brain memorises and processes complex tasks.
- F. Ajejas et al., "Unraveling Dzyaloshinskii–Moriya interaction and chiral nature of graphene/cobalt interface." Nano Letters, 18 (9), 5364-5372 (2018)
- F. Ajejas et al., "Thermally Activated Processes for Ferromagnet Intercalation in Graphene-Heavy Metal Interfaces." ACS Applied Materials & Interfaces, 12 (3), 4088-4096 (2019)
- P. Olleros-Rodríguez et al., "Intrinsic mixed Bloch-Néel character and chirality of skyrmions in asymmetric epitaxial trilayer." ACS Applied Materials & Interfaces, 12 (22), 25419-25427 (2020)