As sunlight filters through a forest canopy, chlorophyll is hard at work capturing the energy of photons. Inspired by nature, researchers at NTNU are working on light-capturing dyes for solar cells to generate electricity.

These aren’t the kind of solar cells you’ll see on the roof of a building. In those silicon solar cells, light hits one of two semiconductor layers and frees up electrons to jump between the layers. It’s the movement of these electrons that creates an electrical current. A dye-sensitised solar cell (DSSC) works in a similar way, but one of the semiconductor layers is replaced with a photosensitive dye that absorbs the light and releases electrons instead.

Dye-sensitised solar cells tend not to be as efficient at converting light into electricity as their silicon counterparts. But they work in low light conditions, and can be transparent and flexible, so are better suited to some applications. To really take full advantage of DSSCs, a research project partially funded by the Research Council of Norway (RCN)* is looking for ways to step up their efficiency. 

In a paper published recently in the journal Dyes and Pigments, NTNU PhD candidate David Moe Almenningen and colleagues, Odd Reidar Gautun, Bård Helge Hoff and Svein Sunde have shown that adding a particular molecule to the dyes can increase its light harvesting properties – though so far the additional light comes at a cost.

To harvest light a dye needs to act as an electron donor and an electron acceptor. “When this molecule is struck by a ray of sunlight, then the electron moves from the electron-rich part to the electron-poor part,” says Almenningen. By adding something in-between the donor and acceptor, chemists are able to increase the amount of light the cell harvests. 

Almenningen’s research is investigating the addition of compounds featuring thiophenes, a molecule similar to benzene but containing sulphur. Thiophenes are electron-rich, so would be expected to increase the light harvesting properties of the dye, he says. And recent experiments show that they do: the dye with the most thiophenes was the one that harvested most light. 

However, it turns out that increasing the amount of light a dye captures doesn’t automatically mean better solar cells. Put simply: you might get more electrons, but they don’t necessarily go where you want them to. 

In his experiments, Almenningen found that though it absorbed the most light, the dye with the most thiophenes actually made the least efficient solar cell. “You think you're doing something brilliant by increasing the light harvesting ability, but then there are other reactions going on in the solar cell that are negatively affected by these modifications,” he says. 

He and his colleagues hope to find a way to avoid those counterproductive effects and take advantage of the improved light collection. Their next step is to try modifying the dye chemically so the electrons can only go in one direction. If this is successful, it could lead to more efficient solar cells.

Finding a way to increase the efficiency of DSSCs is one of the roadblocks to widespread use. The current highest efficiency is around 12%, compared with closer to 20% for a traditional commercial silicon solar cell. 

If researchers are able to harness the light captured by dyes in these solar cells more effectively, DSSCs would potentially offer an advantage over traditional crystalline solar cells when it comes to scaling up: they are cheap to make, because they don’t need a clean room or vacuum technology. 

One promising avenue for DSSCs would be to integrate them into buildings to capture the dimmer light that is typically found indoors. “That’s where these solar cells shine,” says Almenningen. “Also they look quite pretty. You can customise any colour you want, they can be see-through.”

For Almenningen, though, the reward is in figuring out how changing the chemical structure affects the performance of the dye: “The chemistry in itself is what's fascinating.”

*RCN project numbers: 262152, 226244 and 295864

Kelly Oakes, March 2021

A list of NTNU papers related to nanoscience, nanotechnology and functional materials published in March 2021 may now found on NTNU Nano's page for publications.

Modern-day computers rely on the fact that electrons have charge. But electrons have another fundamental property called spin – a measure of magnetic orientation – that researchers hope to harness to create a new generation of computer chips. Spintronics – short for spin transport electronics – could lead to faster, more stable, and less power-hungry devices. 

An electron’s spin is a bit like a compass needle that points north or south. Magnetic hard drives already use the spin of electrons to store information in the form of binary 0s and 1s, which your computer can then translate back into human-readable information. But traditional computer processing ignores spin entirely. Using spin for computation would mean processing and storage could happen on the same chip.

 In most materials, there are equal numbers of electrons with   spins that point in opposite directions, so from the outside   they all appear to cancel out. These materials are known as   antiferromagnetic, and Thomas Tybell, a professor in the   department of electronic systems at NTNU and his colleagues   are looking for ways to engineer them for use in future   spintronic devices. “If we use antiferromagnetic materials,   where the spins cancel, they are very robust against   perturbations,” he says.

That stability is a big plus. Let’s say you’re working on an important document, and just before you hit save there’s a power cut. With conventional computing, you have probably just lost your work. But the spin of an electron stays the same even when the power is lost, so on a spintronic computer your work would be preserved. 

But to create spintronic devices, we first need materials that allow us to reliably control spin. 

One big challenge is engineering materials without internal boundaries that could mess with the spin of electrons and result in lost information. These boundaries  – called domain walls – occur where the repeating pattern of atoms in the material doesn’t quite match up. 

Recently, Tybell and his colleagues have found a way to make thin films from antiferromagnetic materials that look like they have no domain walls at all. By changing the arrangement of atoms – known as the lattice – in the substrate onto which the thin film is deposited, they can ensure the crystal grows in such a way as to avoid creating those internal boundaries. “Our key to controlling the physical properties is the lattice,” says Tybell.
“This is just a grey, boring image but actually what’s important is that each pixel has the same magnetic axis,” he says. “Suddenly you have no domain walls.”

In recent years other researchers have shown that it’s possible to create a single crystal that doesn’t have domain walls. This new work shows it is possible in thin films, too. “That’s important because suddenly you have unprecedented new possibilities for devices,” says Tybell. “If you want to make a device that works you can’t work on a single crystal, you have to make thin films.” 

There are still a number of challenges to overcome before spintronics goes mainstream, and Tybell says he can’t be sure how long it will be until you’ll be able to hold an entirely spintronic computer in your hand. “It will depend on how well we can control the materials to allow them to be mass produced,” he says. “I hope it’s soon, but I fear it’s quite in the future.”

But, he points out, the concept behind the transistors that are present in every computer around the world was first patented in 1925. It then took two decades until the first working transistor was realised by researchers working Bell Labs in the US, and several more years until they were in widespread use.

In the meantime, the materials Tybell and his colleagues are developing will not go to waste: they can also be used by researchers studying quantum objects from a fundamental physics point of view. And while it’s not easy to predict where that could lead, there’s always a chance it might prove vital in the future, one way or another. 

“We should not forget that if you can grow single crystalline thin films, it opens up new avenues to study quantum phenomena and learn about these materials in a way that might be important for quantum technologies in the future,” he says. “I am sure there are many things still to discover.”

Kelly Oakes, February 2021
 

A list of NTNU papers related to nanoscience, nanotechnology and functional materials published in February 2021 may now found on NTNU Nano's page for publications.

This week's news articles describe the story behind the development of NTNU’s COVID-19 test, the new Norwegian Quantum Computing Centre, and how advances in silicon processing may lead to more environmentally friendly solar cells .

A list of NTNU papers related to nanoscience, nanotechnology and functional materials published in January 2021 may now found on NTNU Nano's page for publications.

Sometimes, when it comes to friction, less is more – at least that’s what several experiments over the last decade seem to have shown in the case of friction caused by layered materials. But it wasn’t until recently that researchers at NTNU figured out what was actually going on.

Friction probably isn’t something you think about on a daily basis, but as anyone who’s ever slipped over on an icy pavement could tell you, it can play a crucial role in many situations. The downside of friction, though, extends far beyond icy pavements: along with wear, it is responsible for approximately 23% of the world’s energy consumption. 

“Friction is a huge technological problem,” says Astrid de Wijn, a professor in the department of mechanical and industrial engineering at NTNU. “In industrialised societies, where we have machines that are moving constantly or very fast, friction is enormously costly.” 

Studying friction is not as simple as classroom physics demonstrations involving a wooden block on a ramp suggest. “What is really happening is that the surfaces are rough and they meet at some points that are typically quite small,” says de Wijn. “When we study friction we are thinking about these contact points and how they behave.” 

In order to really understand what they can see happening at the real world macroscale, researchers need to be able to explain the complex behavior of the materials at the nanoscale. “Many different things are happening at different length and time scales, and it makes friction very interesting,” says de Wijn.

Layered materials  – such as graphene, which is a single layer of carbon atoms arranged in a honeycomb pattern – generally have low friction. They are already used in lubricants, but learning more about how they work could enable us to make the world’s machinery run more smoothly and reduce our energy bill on a global scale. 

But in the last decade, researchers studying how friction works in materials like graphene have found that a single layer actually creates more friction than several layers. “People didn't understand that, and for years they were struggling with it,” says de Wijn. “They did simulations and they reproduced this behaviour but couldn't figure out what was really happening.” 

The problem was, while there were plenty of results apparently showing what was causing the material to behave like this in particular situations, the explanations proposed by different researchers seemed to contradict each other, and nothing stood out. “They all had good arguments and evidence that, in their system, it was their suggested mechanism that was doing it, says de Wijn.

In a recent paper published in Nature Communications, de Wijn and PhD student David Andersson solved the problem. It turns out that all of the proposed solutions are, in a way, right. 

One existing model that often proves helpful for understanding friction was first proposed in 1928 and consists of three elements: a support, a spring and a tip. The friction is then the force required to pull the tip across a sheet. While that works well for explaining many situations, it falls apart when layered materials are involved. So de Wijn and Andersson added just one variable to describe what is happening inside the layers of the sheet that the tip is being pulled across. “We didn't specify what that variable meant exactly – if it was a scrunching up of some kind, or some bending or one of the many possible things that people had proposed,” she says. 

That simple tweak turned out to be the key to explaining several previous results, both from real world experiments and computational models. “Suddenly all the pieces fell into place and we understood what was happening,” says de Wijn. “It could be different mechanisms giving rise to the same kind of dynamics.”

Unfortunately, because of travel restrictions due to the coronavirus pandemic, de Wijn and Andersson have not been able to present the work at many conferences or discuss it with colleagues as widely as they would otherwise have done – at least not in person. Nevertheless, now this puzzle has finally been solved, it opens up new avenues for investigating how friction works in layered materials, and could pave the way for new technology to reduce it. 

The work is not over yet, though. The next step for de Wijn is to figure out how thermal fluctuations affect the system. “It's not just an academic puzzle for us,” she says. “Solving this means that we are one step closer to making friction lower.”
 

Kelly Oaks, January 2021