CORDIS - EU research results

Spin Orbitronics for Electronic Technologies

Periodic Reporting for period 4 - SORBET (Spin Orbitronics for Electronic Technologies)

Reporting period: 2020-05-01 to 2020-10-31

Today we live in a digital world where digital data provides the underpinning of much of our life. Many processes both at home and in industry rely on digital data, the vast majority of which is stored in the cloud in magnetic disk drives, a technology that is more than a half a century old. This technology has evolved by ever-decreasing the size of the elementary magnetic data bits but it is generally regarded that this process has reached an end and a new technology is needed. The technology that has the potential to replace magnetic disk drives is Racetrack Memory. This technology invented by the PI of SORBET relies on recent advances in the field of spintronics. The basic concept is that digital data is encoded in magnetic domain walls, tiny nanoscopic objects that separate regions of magnetic material that have their magnetizations directed along opposite directions. These nano-objects are shifted along magnetic racetracks – tiny magnetic nano-wires – by current pulses. A series of domain walls are shifted synchronously backwards and forwards so that these data bits can be read or written at individual reading and writing devices built into or alongside the racetracks. Racetrack Memory has the potential to have enormous storage capacity beyond that of any solid state memory today and rivalling or exceeding that provided by magnetic disk drives. Moreover, since there are no moving parts, unlike disk drives, Racetrack Memory will be much more reliable and will consume vastly less energy.

A main objective of SORBET was to explore the fundamental physics of current induced motion of domain walls so as to find new materials and structures that could allow for more efficient and reliable domain wall motion. A second main objective was to explore means of creating nano-scopic racetracks that could be oriented perpendicular to the substrate on which they are formed. Finally, a third objective was to explore topological nanoscopic magnetic objects that could be used for storing digital data beyond domain walls. All of these objectives were met in SORBET. New efficient methods of manipulating domain walls via spin-orbit and giant exchange torques were discovered. A novel multi-photon lithographic technique was developed to allow for complex 3D racetracks to be fabricated. Anti-skrymions and elliptical Bloch skrymions, two novel and distinct topological non-collinear spin textures, were discovered in a family of inverse tetragonal Heusler compounds. The results of SORBET have pushed Racetrack Memory to the cusp of technological application.
SORBET concerns the exploration of novel devices that rely on spin-orbitronics. SORBET especially focuses on Magnetic Racetrack Memory in which a series of magnetic domain walls are moved to and fro along magnetic nanowires using nanosecond long current pulses. The presence or absence of a domain wall correspond to a digital bit of information. If the domain walls can be stored very densely in tiny magnetic nanowires – the racetracks - and moved with current pulses at high speed, then it is possible to create an entirely new form of non-volatile memory with a very high density and a very high performance. To date the nanowire racetracks have been formed on flat surfaces: one objective of SORBET is to develop techniques to enable 3D racetracks. Other objectives of SORBET include studying the underlying physics of the current induced domain wall motion to enable more efficient domain wall motion. Another important objective of SORBET is to explore alternatives to domain walls such as anti-skyrmions that are tiny magnetic nano-objects.

SORBET met or exceeded all of its objectives. Major results included novel materials that include ultrathin dusting layers that reduced the critical current density needed to move the domain walls and enabled a five-fold increase in domain wall speed for otherwise the same current density. Another result was the discovery of a novel chemical templating technique that allow for the deposition by sputtering techniques of thin films of Heusler compounds that are only 1 unit cell thick. Ultra-thin films are essential for technological applications of such materials for racetrack and other applications such as magnetic random access memory. An important series of experiments that goes beyond the original goals of SORBET relate to fundamental aspects of “chirality”. One highlight in SORBET was the exploration of the interaction of chiral organic molecules with perpendicularly magnetized thin films, similar to those used to form racetracks. This work, in collaboration with colleagues from the Weizmann Institute and Hebrew University, Israel, was published in a paper in Science. During SORBET an advanced multi-photon lithographic capability for building 3D nano-structures was built. This enables 3D freeform fabrication through a two-photon polymerization initiation that is localized in space. The replication of a 3D model is achieved by moving the focus of the laser beam inside a viscous liquid, controlling the places where the polymerization occurs in space. We have constructed complex 3D racetracks with chiral forms in which we have demonstrated the current induced motion of chiral domain walls depends on their chirality.
We discovered anti-skyrmions, a novel magnetic nano-object, that has unique properties. We demonstrated that the size of anti-skyrmions can be tuned from a few tens of nanometers to microns in the same material, which was not anticipated by any theoretical model. We also showed that anti-skyrmions are intrinsically stable as compared to other non-collinear spin textures.

We discovered for the first time that one magnetic material can sustain two distinct topological non-collinear spin textures. The Heusler compound Mn1.4PtSn in which we had previously discovered anti-skyrmions, can also support “elliptical Bloch skyrmions”. These are unusual nano-objects for which the major axis of the ellipse is oriented along one of two crystallographic directions in the crystal (100 and 010). Along these direction Bloch walls are favored which minimize magnetic-dipole interactions. We also fabricated racetracks with nano-scale dimensions and demonstrated the formation of a single row of nano-objects that can be either anti-skyrmions or Bloch skyrmions. We proposed that these distinct nano-objects could be used to store “0”s and “1”s for racetrack memory devices with increased stability of the magnetic bits compared with conventional racetrack memory devices in which “0” and “1”s are stored as the presence or absence of chiral domain walls.

We explored the properties of thin films formed from chiral antiferromagnetic structures namely hexagonal Mn3X (X=Sn, Ge, Sb). We developed methods to grow high quality films by sputter deposition and explored the anomalous transport properties of these films. We also used magnetic dichroism techniques at the BESSY synchrotron to explore the field and temperature dependent magnetization of these compounds. Beyond SORBET we have recently demonstrated a novel method for manipulating the antiferromagnetic ground state of these films using what we have termed “Seeded Spin Orbit Torque”. This method allows the manipulation of the state of films which are very thick and which, therefore, can be highly thermally stable.
Size tunability of magnetic anti-skyrmions in a wedge‐shaped lamella of Mn1.4PtSn
Magnetic force microscopy image of anti-skyrmions in Mn1.4PtSn