8 April, 2019
Could bio-inspired self-assembled magnetic structures make computers more efficient?
Anyone who has ever picked up a phone while it’s charging, or worked with their laptop on their knee, can tell you that electronic devices produce heat. It’s all down to the way electrons push through the device’s components, continuously giving up small amounts of heat energy. But those interactions add up, and for the server farms that the internet relies on, this ‘waste’ consumes vast amounts of electricity, and costs billions of dollars to manage. Some organisations are trialling systems to recycle this heat, but wouldn’t it be better if our electronics didn’t waste so much energy in the first place?
That’s the view of newly-promoted MacDiarmid Institute Principal Investigators Dr Jenny Malmström (University of Auckland) and Dr Simon Granville (Victoria University of Wellington). They’re collaborating on an ambitious project to build a new generation of devices from the bottom up, with self-assembled magnetic nanoparticles.
It’s based on the idea that, rather than using an electric current to send information, computers could be driven by spin waves – tiny oscillations in the magnetic properties of certain materials. This field of research, called magnonics, only emerged in the last decade, but ambitions are already running high. “The potential gains are enormous,” says Dr Granville. “Magnonic devices could theoretically use less power, and waste less energy, than conventional electronics, while also operating at speeds way beyond anything we have today.”
They could do this largely because magnonic devices don’t rely on a flow of electrons to operate. The ‘current’ is actually more like a wobble in the magnetic field that propagates through a material, and because it occurs without any particle motion, it doesn’t generate heat. Magnonics could also help make individual components much faster, by shrinking them to unimaginably small sizes. “The ability of chip manufacturers to continuously improve computing power by shrinking transistors is finally being hampered by the laws of physics,” says Dr Granville. “Once they drop below seven nanometres, transistors start becoming really unpredictable”. Magnonic devices, however, would only be limited by the gaps between atoms, so if they could be made small enough, it would be possible to push them harder and faster than traditional transistors.
And that ‘if’ is where Dr Malmström comes in. “Even today’s very best fabrication facilities can’t pattern magnetic materials into the sorts of patterns we’d need to make magnonic devices practical,” she says. So Dr Malmström and her team have been applying lessons from both biology and chemistry to the problem. “We wanted to focus on different self-assembly approaches because it could give us a low-cost way to produce very small patterns over large areas,” Dr Malmström explains.
They started by adding iron salts into materials called block copolymers, which, because of their chemical structure, naturally assemble into nanopatterns. They’re also now collaborating with MacDiarmid Institute Principal Investigator and University of Auckland’s Professor Penny Brothers and her postdoc, Dr Seong Nam, to put polyoxometalates – molecular clusters with a magnetic core – into their self- assembled materials.
In addition, Dr Malmström recently published a paper with a group of colleagues that included Auckland-based Associate Investigator, Dr Laura Domigan, and PhD student Sesha Manuguri. In it, they report on the use of proteins as a way to organise nanoparticles into specific patterns. By trapping particles in the centre of donut-shaped protein structures, and using self- assembly to stack them, they’ve shown that they may be able to create nanoscale wires. “This seems unbearably cool to me,” says Dr Granville. “As a physicist, I’m used to looking at things from the top down, but here, we’re building these structures from the bottom-up!”
Dr Granville and Dr Malmström’s cross-discipline approach to this challenge is what makes it unique, and they’ve used it to build a large team that spans the MacDiarmid Institute. “Emeritus Investigator Professor David Williams was the first person to join the dots,” says Dr Malmström. “He’s a fantastic ‘big ideas’ man, and understands the value of collaboration. Former MacDiarmid Institute Principal Investigator, Juliet Gerrard – now the Prime Minister’s Chief Science Advisor – has also been central to this work.” Other Principal Investigators involved include Auckland’s Associate Professor Duncan McGillivray, Professor Alison Downard from the University of Canterbury, and and former MacDiarmid Institute Director Professor Thomas Nann.
An important addition to the team came last year, when Sesha Manuguri joined the University of Auckland as a postdoc, supported by the MacDiarmid Institute. “Sesha’s work sits perfectly between what we’re doing in functional nanostructures, and what Simon is doing in thin-film magnetic materials, so she’s been a fantastic addition” says Dr Malmström. Dr van der Heijden recently joined Dr Granville on a research trip to China, where they were joined by Manuguri and another PhD student, Kyle Webster. There, they met and worked with a team of researchers at Beihang University. “I’ve known Professor Haiming Yu since he was a PhD student in Switzerland,” says Dr Granville. “He now runs a unique facility that can be used to measure the performance of exactly the type of magnetic materials we’re trying to make.”
The visit itself was very positive, giving the team the first round of results on their samples, some of which were described as “unexplained but exciting”, by Manuguri. It’s a sentiment echoed by Dr Granville, too. “With the first spin wave measurements, we knew not to expect a ‘smoking gun’, but there were definitely lots of positive things to look at,” he says. “The visit also gave us the opportunity to sit down with Haiming and plan the next steps.”
And they have their eyes on a very big prize – getting closer to building working magnonic transistors, and testing their performance. “Simon, David and I have talked about making real devices,” says Dr Malmström. “No-one else has managed to build even the simplest logic gate, so yes, it’s ambitious.
But I think we have the right team to tackle it, so why not try?” With the world searching for an alternative to today’s computers, it’s a timely goal to aim for.