1.2 Taking the heat off – Annual Report 2017
Each time you type into Google and hit ‘search’, somewhere in the world a data centre responds. This data centre – a repository for masses of information – will be one of many worldwide (there are even four in New Zealand). Globally, data centres use almost ten times as much electricity per year as the whole of New Zealand, and their collective energy use is projected to treble in the next decade.
It turns out not all electricity used by data centres goes into running your search query. A good chunk of it is used to cool down the computers.
It’s this chunk that two MacDiarmid Institute teams are addressing, in quite different ways.
One of the MacDiarmid teams is a group of physicists at Victoria University of Wellington working with rare earth nitrides. The researchers have combined these rare earth nitrides with super- conducting electronic materials to make new superconductors, and already have a couple of patents under their belts.
The other is a multidisciplinary team led by MacDiarmid investigators at the University of Auckland and involving our investigators at Victoria University of Wellington and the University of Canterbury. Using biology, this team is looking at new ways to assemble nanowires and nanoarrays for tiny transistors. The team brings together researchers with deep knowledge in specialist areas and is using this cross-discipline expertise to create potentially game-changing technology.
Working to reduce electricity use 100-fold
Principal Investigator and Victoria University of Wellington Associate Professor Ben Ruck says that data centres use electricity in two ways. “When you run big data centres, you pay twice for electricity. You pay to run the computer, and you pay to get rid of the heat it creates. Heat is such a big deal that they actually build data centres in Sweden, where they can just keep the doors open.”
Along with the VUW MacDiarmid team, Associate Professor Ruck has found and patented materials that could cut the amount of global electricity used by data centres by 100-fold, from 3 percent to 0.03 percent. “We’ve been working with rare earth nitrides like samarium or gadolinium nitride – simple compounds that act as magnetic nanomaterials – and combining these with superconducting electronics, based on materials such as niobium. To our surprise, we found that samarium nitride is a superconductor itself,” explains Associate Professor Ruck.
Superconductivity in samarium nitride is unexpected because it is a magnetic semiconductor and superconductivity and magnetism are usually incompatible bedfellows.
For example, in samarium nitride the electrons organise themselves into a state where their so-called intrinsic ‘spins’ all point in the same direction. This makes it magnetic. By contrast, superconductors such as niobium almost always have electrons in pairs with opposite spin directions. Somehow, samarium nitride finds a way to accommodate both types of ordering.
The team (which also includes MacDiarmid Principal Investigators Dr Franck Natali, Emeritus Professor Joe Trodahl, Professor Michele Governale and Professor Uli Zuelicke, as well as Associate Investigator Dr Simon Granville, and more recently Emeritus Investigator Professor Bob Buckley) has been patenting as they go. They have two patents already and more on the way, with 2017 KiwiNet funding to help.
Associate Professor Ruck says that while they’re still some way off commercialising it, there are good prospects for technology like this to one day be manufactured in New Zealand. “These components are tiny and light and high value. We could ship them anywhere.” The VUW MacDiarmid team is already collaborating with a US company to develop their ideas further.
Associate Professor Ruck says the project is an example of a long-term deep-research project, funded over many years by a Centre of Research Excellence, starting to show its usefulness in the practical world.
“When we started this MacDiarmid project in 2006, we were trying to understand the basic properties of the materials. From about 2012, we could then start to look at how these materials could be used to make computer memory. And from 2015, we’ve been able to tie it in with superconductivity, to not only make computers faster, but to reduce the heat and the big impact these data centres are making on the environment.“
“Ten years into this research, our MacDiarmid team is still the world leader in this area, and because of our patents, we now have international companies wanting to work with us.“
A multidisciplinary approach
Another MacDiarmid team is approaching the heat problem in a different way.
Sitting together at a MacDiarmid Institute Functional Nano- structures theme meeting in mid 2016, Victoria University of Wellington physicist and Associate Investigator Dr Simon Granville, and University of Auckland MacDiarmid Principal Investigators biochemist Professor Juliet Gerrard, chemist Professor David Williams, and materials engineer and Associate Investigator Dr Jenny Malmström were pondering this same problem – how to reduce the heat produced by computers in general, and data centres in particular. By working across their four disciplines, they’ve come up with an entirely new approach.
Dr Simon Granville works with magnonics – studying magnetic materials able to generate magnons, or spin waves, that could be downsized to the level of modern transistors – making devices that will be very fast while saving a lot of energy. In theory, magnonics is a good idea for a technology to replace electronics.
However, the building blocks are so tiny (much smaller than 100 nm) that assembly proves a huge challenge. The multidisciplinary MacDiarmid team put their heads together to come up with a couple of cunning ideas.
A biological solution to an electronic problem
The first idea was to try using protein ‘doughnuts’ (a regular building block of a more manageable size – 15 nm diameter, with a hole in the middle of about 8 nm) to assemble nanowire and nano surfaces. This method had already been used to assemble other nanoparticles by Principal Investigator and biochemist Professor Gerrard.
MacDiarmid PhD student Sesha Manuguri – jointly supervised by Professor Williams and Dr Malmström – has since managed to restrain iron nanoparticles to the 3-4 nm dimensions of the inside of the protein doughnut, by making the particles assemble inside the doughnuts.
These protein doughnuts can stack into long tubes. Dr Malmstr ö m says the aim is for each doughnut to be a carrier for magnetic nanoparticles, and create long magnetic nanowires.
New MacDiarmid Institute Associate Investigator and University of Auckland biochemist Dr Laura Domigan (who collaborates with Professor Gerrard) is working to place single protein doughnuts in an organised pattern on a surface. If doughnuts can be placed on a surface they could be used to hold nanoparticles in an organised pattern. Dr Domigan is currently working on assembling them on surfaces.
Using polymers to assemble thin nanoscale layers
The second idea hatched at the functional nanostructures theme meeting involves using polymers (another name for plastics) to assemble nanoparticles. By using a so-called block copolymer – with two parts that don’t like each other, much like oil and water – the polymer can organise itself into regular nanopatterns.
These polymer patterns can be used to make an ordered pattern of nanoparticles, although the particles that can be made in the polymers are larger than those made in the protein doughnuts. There is also the potential to use the polymers themselves to order the protein doughnuts.
It is then up to Dr Granville and his collaborators to test these materials for magnonic properties, by measuring the wavelengths of the spin waves produced using the patterns.
“If spin waves can be excited to feature wavelengths in the nanometre range, it might allow for the downscaling of devices that work at gigahertz frequencies and can compete with the nanosized transistors of today“, he says.
Dr Granville says what excited him was connecting people across a multitude of disciplines – biochemistry, chemistry, materials engineering, and physics.
“This is exactly the sort of project MacDiarmid is good at. It’s hard to see another organisation managing it.“
“We’re bringing together people with a deep knowledge in specialist areas and seeing how magnonics can be applied to be completely revolutionary. That’s the goal of the project“, says Dr Granville.
Other than potentially creating a revolutionary technology in the form of magnonic transistors, Dr Malmström says there are other valuable benefits to such a project.
“This is another thing that excites me. You are doing something so far out there, that’s where the big breakthroughs come from. We may have built up an entirely new science.“
Professor David Williams says this kind of research is really only possible in a Centre of Research Excellence (CoRE).
“Only a CoRE would naturally facilitate a group as diverse as this to come together to hatch ideas.“ He loves the challenge. “You want your brain to hurt, to get out of your comfort zone. You attract people when you do things like this. You respond to sparks and generate your own sparks. We do the weird stuff no one else is doing.“
A newly created MacDiarmid postdoctoral position will slot into the middle, providing the glue between the investigators, and enabling functional measurements of the new materials.
Other MacDiarmid Institute researchers involved include Canterbury University Professor Alison Downard and University of Auckland Professor Penny Brothers and Associate Professor Duncan McGillivray (who will run various tests, including neutron scattering); a PhD student who is putting gold particles in the middle of doughnut protein; and another PhD student working with other polymers to form an organised nanoscale film.
One day, with these new technologies, perhaps we’ll be able to ‘like’ our friend’s Facebook post without collectively putting such a strain on global energy use or contributing to greenhouse gas emissions.