Multi-junction solar cells come from just one compound

By growing or annealing zinc tin phosphide (ZnSnP₂) layers at certain temperatures they were able to introduce disorder into the structure that allowed different band gaps to be created.

‘There are different ways that this could be used, one is to try to enhance the performance of a silicon solar cell, but another would be to have a thin-film solar cell,’ said Aron Walsh of Bath University. ‘The two main thin-film technologies are cadmium telluride and CIGS [copper indium gallium selenide] — this would be an alternative, so you could have a low-cost thin-film cell.’

Multi-junction solar cells use combinations of exotic rare materials such as germanium, gallium and indium, each of which display a characteristic band gap that can capture a specific part of the solar spectrum. Because of this, less energy is lost and so conversion efficiencies are very high.

However, due to their great expense, multi-junction cells have only really found a market in the space industry, to power satellites, for example, and as tiny chips in concentrated photovoltaic devices. As a consequence, attention has turned to cheaper more readily available materials.

‘I’ve been working on zinc tin compounds probably for the last five years, and they are just two elements that seem to work very well for solar cells and especially in the UK there’s a natural abundance of zinc and tin, but they haven’t really been commercially exploited yet,’ Walsh said.

Walsh, along with a colleague at University College London, found that growing or annealing ZnSnP₂ at different temperatures resulted in a change in structure. Crucially at around 990K there is a transition at from a so-called chalcopyrite structure to a more disordered sphalerite state.

In this disordered state, the material exhibits electrical band gaps as low as 0.75eV, which can pick up on some of the low-energy portions of the solar spectrum such as infrared photons, which traditional single-junction solar cells cannot harness.

When studying different levels of disorder in their ZnSnP₂, the team found that although aspects of the structure become more disordered, the lattice constant actually remains the same. The upshot of this is that they can effectively bolt on different layers on top of each other to create a multi-junction device.

The team will now focus on fabricating some prototype cells, the challenge being to introduce the varying temperature regime in a scaleable way. For this, Walsh says burgeoning techniques such as atomic layer deposition could be adapted.

‘Having a low-cost triple-junction solar cell, that’s what seems to work well,’ said Walsh. ‘If you have too many junctions the light can’t pass through all the way, that’s when you have to move to concentrated sunlight. So if you want a solar cell that’s going to work outside and it might be a little bit cloudy then you can’t have too many layers.’