56 57 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 1 / 2 2 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 1 / 2 2 Magnetic whirls join the race for new computing paradigms Related publication: Jani, H., Lin, J.-C., Chen, J., Harrison, J., Maccherozzi, F., Schad, J., Prakash, S., Eom, C.-B., Ariando, A., Venkatesan, T., & Radaelli, P. G. Antiferromagnetic half-skyrmions and bimerons at room temperature. Nature 590 , 74–79 (2021). DOI: 10.1038/s41586-021-03219-6 Publication keywords: Magnetism; Spintronics,; Information technology R acetrack non-volatile memory, developed by IBM, is one of the most innovative information technology concepts to emerge in the past two decades. It works by creating tiny ‘disturbances’ in an otherwise uniformmagnetic medium, stored and retrieved bymoving them along a track. The difficulty of stabilising the disturbances and propelling them at high speed has prevented the racetrack concept from reaching the market. The original racetrack concept employed ferromagnets – materials in which the magnetic atoms all point in the same direction. However, recent research has shown that disturbances can travel much faster in antiferromagnets (with the magnetic atoms pointing in opposite directions). In a previous investigation at Diamond Light Source, an international team of researchers discovered that the antiferromagnetic material α-Fe 2 O 3 supports a family of vortex-like disturbances (known as antiferromagnetic textures) that are potentially very stable. The team is now investigating how to gain full control over the formation of these textures. Antiferromagnetic textures are much more difficult to visualise than their ferromagnetic counterparts, requiring the powerful imaging techniques available at Diamond. The Nanoscience beamline (I06) delivers full control over the X-ray beam properties, allowing the researchers to map the antiferromagnetic textures in exquisite detail. Using X-ray Photo-Emission Electron Microscopy (X-PEEM), they acquired images of these textures over a field of view of a fewmicrons. a-Fe 2 O 3 is under the spotlight for application in fast and energy-efficient computing and being able to create magnetic textures in this material easily is extremely exciting. Understanding the formation and control of such textures and their potential application in energy efficient computing represents a significant milestone in the field of information and communication technology . The information and communication technology ecosystem has been transformed over the last 60 years due to the rapid density and cost scaling of charge-based silicon transistors, a phenomenon popularly dubbed as the ‘ Moore’s Law ’. While this has increased computer accessibility worldwide, with ~10 billion connected devices, it consumes a lot of energy and contributes more than 2% of the world emissions (on par with aviation). With the rise of Big Data, Artificial Intelligence, Internet of Things (IoT) andmobile-computing, the overall computing energy demand is expected to a rise exponentially. This has resulted in a quest for novel 'beyond-Moore' computing paradigms, which explore energy-efficient ‘normally-off’ non-volatile memory, combined with unconventional processing architectures (such as logic-in-memory or brain- inspired reservoir computing). One of the most innovative information technology concepts that emerged in the past two decades is the so-called racetrack memory , proposed by IBM 1 in the mid-2000’s. This involved the use of magnetic patterns (also known as ‘magnetic textures’) instead of charges to store information bits (‘0’ and ‘1’). These textures remain stable even in the absence of external power, making them highly energy efficient. Typically, one creates tiny textures (such as magnetic walls or ‘whirls’) in an otherwise uniform magnetic medium, which can be stored by ‘parking’ them along a magnetic track and retrieved by pushing them along the same track when needed. This racetrack concept has not reached the market yet, largely due to the difficulty of stabilising the textures and propelling them at high speed. To make progress, a new concept was needed. The original racetrack concept employed ferromagnets –materials inwhich magnetic atoms all point in the same direction. Recent research has instead focused on antiferromagnets , which are made of alternate atoms pointing in opposite directions (N/S/N, E/W/E etc.). This special arrangement makes antiferromagnets very stable and robust. Moreover, it has been theoretically predicted that textures in such materials ( e.g. , antiferromagnetic whirls), would also be very scalable down to small sizes and would travel at ultra-fast speeds (up to few km/s) – making them ideal to build dense, fast and energy-efficient racetracks. While the alternating atomic arrangement in antiferromagnets is highly beneficial, it also makes the detection and control of tiny antiferromagnetic textures very difficult via standard magnetic techniques. These are the two major barriers hindering experimental progress in this field. The recent study published in Nature 2 was aimed at addressing these very issues. Firstly, to visualise antiferromagnetic textures the authors used the powerful X-ray microscope at the I06 Beamline to perform X-ray Photo-Electron Emission Microscopy (X-PEEM). The X-PEEM technique generates images of magnetic textures by collecting electrons emitted following the absorption of carefully polarised X-ray beams. Building on previous work 3 , they exploited the full control over the X-ray beamproperties and sample orientation, allowing tomap out the antiferromagnetic textures in exquisite nanoscopic detail (Fig. 1). Secondly, to control the formation of antiferromagnetic whirls, the authors focussed on the antiferromagnet α-Fe 2 O 3 , which is extremely abundant and cheap (it is a main constituent of rust). Having discovered in a previous investigation at Diamond that α-Fe 2 O 3 supports a family of vortex-like textures that are potentially very stable 3 , the next step was to gain reproducible control over the formation and evolution of these textures. To realise this, the authors drew inspiration from a celebrated idea in cosmological physics, from nearly 50 years ago, developed by the British physicist Sir Tom Kibble 4 . He proposed that a phase transition in the early universe, during the cooling after the Big Bang, may have resulted in the formation of cosmic whirls. They hypothesised that the same phenomenon could be exploited to create antiferromagnetic whirls by driving α-Fe 2 O 3 across its magnetic phase transition (Fig. 2). Experimentally, this involved heating ultra-high quality α-Fe 2 O 3 thin films though a particular temperature, which can be conveniently chosen by slight chemical or physical alterations 2,5 . I06 images in this study confirmed that they were able to create not one but a veritable‘zoo’ of whirling magnetic textures (Fig. 1). Among these, a particular variant known as bimeron appears to display specific ‘twists’which would make it highly stable and suitable for racetrack applications. Moreover, other textures, such as merons and antimerons , could be employed in novel computing architectures inspired by the human brain. Crucially, all these textures were found to be stable at room temperature and creatable reproducibly by cycling through the magnetic phase transition. Given that α-Fe 2 O 3 has been under the spotlight for application in fast and energy efficient computing for some time, this discovery of a facile and reversible pathway to create a wide family of antiferromagnetic textures in this material is extremely exciting. The next steps are to design proof of principle devices to realise ultra-fast electrical control of the whirling antiferromagnetic family, taking the work a step closer to realistic racetracks. References: 1. Parkin, S. S. P. et al. Magnetic domain-wall racetrack memory. Science 320 , 190–194 (2008). DOI : 10.1126/science.1145799 2. Jani, H. et al. Antiferromagnetic half-skyrmions and bimerons at room temperature. Nature 590 , 74–79 (2021). DOI: 10.1038/s41586-021-03219- 6 3. Chmiel, F. P. et al. Observation of magnetic vortex pairs at room temperature in a planar α-Fe2O3/Co heterostructure. Nature Materials 17 , 581–585 (2018).DOI: 10.1038/s41563-018-0101-x 4. Kibble, T.W. B. (1976). Topology of cosmic domains and strings. Journal of Physics A: Mathematical and General 9 , 1387–1398. DOI: 10.1088/0305- 4470/9/8/029 5. Jani, H. et al. Reversible hydrogen control of antiferromagnetic anisotropy in α-Fe2O3. Nature Communications , 12 , 1668 (2021). DOI: 10.1038/s41467- 021-21807-y Funding acknowledgement: The work done at the University of Oxford was funded by EPSRC grant no. EP/ M2020517/1. The work at the National University of Singapore was supported by the National Research Foundation under the Competitive Research Program (NRF2015NRF-CRP001-015) and by the Agency for Science, Technology and Research under the Advanced Manufacturing and Engineering Individual Research Grant (AME-IRG-A1983c0034). Corresponding authors: Prof Paolo Radaelli, University of Oxford,
[email protected]; Dr Hariom Jani, National University of Singapore,
[email protected] Figure 2: Artist’s impression of whirling textures in a-Fe 2 O 3 racetracks created after performing a magnetic transition analogous to the Big Bang cooling. MagneticMaterials Group Beamline I06 Figure 1: (Centre) X-PEEM vector map of antiferromagnetic textures at room temperature in a-Fe 2 O 3 . Red-Green-Blue colours represent the axis of the antiferromagnetic directions (Adapted from Ref 2). The four insets (left and right) show the schematics of magnetic arrangements in merons (blue circles and white squares), antimerons (yellow circles) and bimerons (dashed black ellipses). Their antiferromagnetic nature is shown schematically as two adjacent and oppositely aligned magnetic layers.