Putting A Spin On Machine Learning, Atom by Atom

There is nothing quite like the magic of magnets. And yet even Richard Feynman, an incredibly gifted science communicator, struggled to explain just how magnetism works. (The video in question is easily found on YouTube. Feynman's slight tetchiness with the interviewer who raises the subject of magnetic forces is not entirely unrelated to the difficulty in explaining their fundamental origin at a level that a non-physicist -- or, indeed, a physicist -- can readily grasp.) Scientists are now at the point, however, where not only can we measure forces on an atom-by-atom basis, but we can harness and exploit those self-same forces to manipulate magnetism right down to the atomic level (and beyond).

The instrument that allows this exquisite level of control of magnetic forces is the scanning probe microscope. A technique that will shortly reach its fortieth birthday, probe microscopy is conceptually rather straight-forward -- its experimental realisation rather less so. An exceptionally sharp tip, terminated in a single atom or molecule, is brought extremely close to a surface such that the tip-surface separation is of the order of the diameter of an atom or less. This atomically sharp probe can then be used in a number of modes to explore, interrogate, and modify the underlying sample surface on an atom-by-atom basis. Some of the most exciting and ground-breaking science ever carried out has involved the scanning probe microscope's unparalleled ability to not only image, but manipulate, matter at the single atom level.

Probe microscopes are not just limited to the imaging and control of atoms; they can go much further. With an appropriately modified tip apex, even the quantum mechanical spin of electrons -- which, ultimately, is the source of magnetism -- is detectable either via the tiny electrical current that flows between the probe and the sample, or, incredibly, via measurement of the minuscule magnetic force between single atoms. Just a couple of months ago (in Oct. 2019), Chris Lutz' group at the IBM Almaden Research Centre reported that they have achieved, in collaboration with researchers in Korea and Oxford, the most precise and coherent control of the spin state of individual atoms ever attempted with SPM. (It's worth noting that IBM is the birthplace of both the scanning probe microscope itself, which was invented by Binnig, Rohrer and co-workers in the Ruschlikon, Zurich research labs, and of SPM-driven single atom manipulation, due to the inspiring efforts of Don Eigler and colleagues at IBM Almaden.)

But the deep, dark secret of the probe microscopist is that a very large percentage of their time is spent coercing and cajoling the probe into providing atomic resolution. Yet even that's not enough -- when that resolution is achieved, the microscopist very often has to maintain the ability to image, move, and spectroscopically interrogate single atoms at the same time, while always being on the look-out for tip-derived artefacts. The component at the core of probe microscopy -- the probe itself -- therefore represents a major, and infuriating, bottleneck in the technique.

This project integrates artificial intelligence, surface science, and nanoscience to take the pain out of probe microscopy. We will develop a machine learning framework that, in essence, "auto focuses" a probe microscope and then takes the SPM to the point where it can learn how to build magnetic nanostructures atom-by-atom and spin-by-spin. By itself. This AI-enabled probe microscope will then be used to carry out a programme of exceptionally challenging experiments whose common theme is the control of magnetism at the most fundamental levels: single domains, single molecules, single atoms, and single spins.

In parallel with its programme of scientific research, the "Putting A Spin On Machine Learning, Atom By Atom" project involves a series of exciting and engaging public engagement activities. The PI has a strong commitment to, and substantial track record in, science communication and engagement in a wide variety of venues and environments spanning YouTube (namely the Sixty Symbols project, a collaboration between the video journalist Brady Haran and the School of Physics & Astronomy at Nottingham, for which he and his colleagues were awarded the IOP's Kelvin Medal in 2016), to the Blue Dot music-science festival (see https://muircheartblog.wordpress.com/2018/07/22/think-graham-norton-meets-the-broom-cupboard-in-space/ ), to "pop sci" writing ("When The Uncertainty Principle Goes To 11", Ben Bella Books (2018)).

Our public engagement programme involves (i) the creation of videos for both the Sixty Symbols and the Computerphile channel, spanning the physics-computer science boundary. (Two videos stemming from the EPSRC Leadership Fellowship I held from 2008 - 2014 have now accrued over 400,000 views due to the popularity of Brady Haran's YouTube channels); (ii) a pop sci book, tentatively titled "F#*ing Magnets, How Do They Work?!", that will use state-of-the-art spin-sensitive scanning probe research to help explain magnetism for a general audience; (iii) the direct involvement of the general public in single atom manipulation via remote (but, of course, highly controlled and tightly restricted) access to the scanning probe microscope; and (iv) a "Magnetic Attractions" installation with the artist, designer, and technologist Matt Woodham.

Although I was a long-standing critic of the research councils' (and, at the time, HEFCE's) so-called "impact agenda", public engagement is clearly recognised by EPSRC as an essential non-academic impact. The "Pathways To Impact" case for the most recent EPSRC project for which I was PI (EP/N02379X/1) was, in fact, exclusively dedicated to public engagement and outreach, as described in a Times Higher Education article written at the time of the award of the grant: https://www.timeshighereducation.com/cn/blog/embracing-impact-seduced-dark-side.

Quoting from that article,

"The research we propose is unashamedly curiosity-driven fundamental science. As such, it is motivated not by the potential for direct short-term economic impact (via, for example, spin-off technology) - and it would be disingenuous of us to suggest otherwise - but by the fascination, importance, and challenges of the underlying science."

This is also true of this fellowship application but close collaborative links with the Nottingham Spintronics group, Diamond Light Source, the project partners (including the company Zyvex, who are focussed on atomically precise manufacturing), and the wider surface science and magnetism communities mean that the research on antiferromagnetic domain dynamics and devices, "twistronics" in 2D materials, and single molecule spin has the potential to impact on a number of industrial sectors -- including, in particular, those for which machine learning is an increasingly important element of their sensor and transducer strategies -- both from the perspective of our "next generation SPM" approach and the intrinsic materials science and device applications that can arise from the work. A key example is the long-standing collaboration between Hitachi Cambridge Lab and the UoN Spintronics group on antiferromagnetic devices, a highly successful interaction that has led to a number of patents.

From the perspective of SPM company involvement, the PI coordinated the 11-partner ACRITAS FP7 network from 2012 - 2016, which included ScientaOmicron, the market leader in ultrahigh vacuum scanning probe systems, along with Bruker, Asylum Research, and a number of other SPM manufacturers. Strategies for IP protection and dissemination are thus well-established.