24 March, 2015
Story by Ruth Beran
When Professor Laurens Molenkamp discovered the quantum spin Hall effect, he knew he was onto something big. At the time he thought it was the type of discovery that an active researcher makes every ten years or so, and with his team, he decided to send it into a bigger journal—Science. “What happened next, we did not really foresee,” he says.
The discovery opened up the whole field of topological insulators. “All of a sudden you notice that it’s not just people from your own field that are interested,” says Molenkamp. “You start seeing better where this theory comes from and that there is a deep connection with topology and with particle theory.”
A topological insulator is a material that behaves like an insulator on the inside but has conducting states on the outside, which means electrons can move along the surface. The ‘very weird metal’ on the outside of all of these materials is peculiar because the electrons behave as elementary particles, like Dirac fermions, says Molenkamp. “That’s the weird thing about it,” he says, “this is a metal but normally we cannot use metals to study that type of physics, but in this metal we can.”
The band structure in these topological insulators means that the direction of movement of electrons is coupled with the spin direction, and carriers (electrons and electron holes) can’t scatter over 180⁰. The result is fantastic mobility (low loss of electrons), which could translate into “big scale devices for the semiconductor community,” says Molenkamp, and superfast computing.
While Molenkamp and his team look at a number of different materials that show this effect, mercury telluride (HgTe) is “the cleanest, at least in terms of the transport physics I do,” he says.
The word topological in the name of these insulators refers to the difference in the structure of the energy bands inside the material as compared with outside. “The band order has to be different from what we normally get,” says Molenkamp. “We call this topologically different, in a mathematical sense.” He gives the example of a ball compared with a mug. A ball can be distorted into a glass without breaking it, but to make a mug from a ball you need to break something to make the handle. It’s topologically different. In the same way, topological insulators have a band structure which is different in a topological sense from other materials.
Molenkamp uses molecular beam epitaxy to grow these topological insulators. “It’s not too much different from very high vacuum evaporation,” he says. “We grow these crystals layer by layer on a substrate of suitable material.”
Leading up to the experimental discovery of the quantum spin Hall effect, which was predicted in 2005, Molenkamp was growing structures called quantum wells with very thin layers, so thin that they were essentially two dimensional. “These wells have very high mobility carriers which have strong orbit effects, so we were playing with them to make spintronic devices,” says Molenkamp. In doing so, he found this topological insulator behaviour. “It was so much more exciting,” he says.
Focusing on HgTe gives Molenkamp and his team a great advantage. “We’re basically the only group in the world who can really grow it for academia,” he says. That wasn’t true 15 years ago when there were 10 or 20 groups doing molecular beam epitaxy in this material system and publishing in the open literature. “Virtually all of them are now doing work for a company or the military making night vision detectors,” he says. Other groups have tried to establish, but safety regulations around mercury are prohibitive. “That gives us an edge,” he says.
Molenkamp and his team are now studying topological insulators for things like exotic superconductivity and Majorana particles (fermions that are their own antiparticle). “Everybody’s going after these Majoranas,” says Molenkamp. “Once we have them, they’re supposed to be good for quantum computing.” These fermions are tricky to find though, because if they do exist they do so in very stable states, rarely interacting, which makes them very difficult to measure. “There is also conceptual difficulties in getting a proper measurement,” he says. He’s also looking at magnetic applications in topological insulators, because of the interesting effects that happen when coupling these materials with a ferromagnet.
Molenkamp was a plenary speaker at AMN-7 in Nelson and was pleased to see chemists there. “That’s something I’ve been missing for a long, long time,” he says. Molenkamp is a chemist by training who “somehow drifted more and more into physics. So I miss the developments that were going in chemistry.” As a physical chemist in the Netherlands, he joined the research lab at Philips Electronics and was given freedom to work with nanolithography and two dimensional gases in gallium arsenide to make semiconductors. “That’s where I learned transport physics,” he says. From there he went to the University of Würzburg where he worked on things like spin injection in semiconductors. The group he runs with a number of other Professors now boasts about 80 people.
During Molenkamp’s first visit to New Zealand for AMN-7, he was surprised to find hardly any groups here working on topological insulators. “This stuff is like the next band wagon to jump on in Asia and America and you see tonnes of groups doing these things, and I was amazed to only see Simon Brown [at the University of Canterbury] doing experiments here, and some activity in Wellington,” he says.
Molenkamp also believes that in a country as small as New Zealand, the MacDiarmid Institute makes a lot of sense. “You have so few universities, you don’t really want to have each faculty specialising in only one thing. Something like the MacDiarmid Institute, where people can collaborate through universities and also from different faculties, I think it’s a very good idea.”
He also acknowledges that the times are over when material physicists can work by themselves. “What I have is really an exception,” says Molenkamp, “and I’m not sure whether this type of thing will continue forever, even in Germany.”