A government-funded project could see steel warehouse and superstore roofs painted with dye-sensitised photovoltaic coatings to turn them into solar power generators.
The three-year project is a collaboration involving the universities of Bath, Bangor and Swansea, Imperial College London and pre-finished steel maker Corus Colors.
According to Dr David Worsely, the director of the engineering doctorate programme at Swansea, the advantage of dye-sensitised semiconductor cells (DSSCs) over silicon cells is their efficiency over a variable light intensity.
'The critical thing about the roofing products is the area,' he said. 'A typical supermarket or warehouse has an area of 20,000 sq m or more, so although DSSCs give relatively low conversion efficiencies overall, you get a reasonable amount of electricity generated.'
The current barriers to DSSCs are ease of manufacture and durability. The team plans to coat or print materials directly on to the steel using Corus' continuous coil coating line, currently used to make building cladding.
Swansea's role is to ensure the steel substrate and its coating are robust enough to take potentially corrosive coating components and ensure that the sun does not actually degrade the paint.
'We'll be looking at optimising the substrate so the steel is not going to be attacked by the corrosive components of the dye-sensitised solar cell,' said Worsley. 'We'll develop barrier layers to prevent the contact of those corrosive components, then carry out accelerated weathering tests to ensure that combined they can last the duration required of them.'
Bath and Imperial both have a long-standing interest in dye cells and will be looking at their construction.
Prof James Durrant of Imperial's photochemistry department said: 'We're using transient spectroscopy to measure rates of electron transfer with different combinations of new materials in the DSSCs.
'We'll try different ways of making and modifying the metal oxide layer and evaluating how that affects device function. For the oxide sensitisers, we will grow organised, oriented titania (titanium dioxide) nanotubes from titanium metal films.'
Prof Laurie Peter of Bath's physical chemistry department said: 'At the moment we are mainly concerned with changing the cell which uses a liquid electrolyte into a cell which uses inorganic or organic solids.
'We're already working on cells which use organic hole conductors, electronically conducting organic materials which replace the liquid electrolyte. Another option is room-temperature ionic liquids, liquid or gels with negligible vapour pressure, so you don't have to worry about evaporation.
'The problem is, it's easy to fill a porous material like the titanium dioxide with a liquid, but more difficult to fill it up with a solid. If you dissolve it and allow the solvent to evaporate, it's quite likely to form crystals inside the pores which don't quite touch all of them. We may use a hybrid solution which is mainly solid but uses some ionic liquid to bridge the gap.
'The final method has to fit in with Corus' manufacturing line, something like screen-printing or a spray-on.'
Bangor is involved with the titania semiconductor layer, which has the sensitised dye attached to the surface to capture more sunlight, improving the efficiency.
Dr Peter Holliman, senior lecturer in inorganic chemistry, said: 'We found you can make a device on a lab scale quite easily, but to make it low cost and to scale it up to a big manufacturing procedure is tricky.
'The titania particles have to be attached to each other to transfer the electrons. We're looking at sintering the titania using flash heating or UV-driven processes to make the particles stick together.'
He said that the team was also looking at the stability of the dyes to get the solar cells to last in the environment for as long as possible. One consideration, he said, was putting absorbent materials into the layers to mop up by-products that might degrade them.
Corus Colors' industrial project manager Dr Maarten Wijdekop said the technology is a long way from commercialisation. 'It's speculative at the moment,' he said, 'but has the potential to make PV surfaces cheaper.'
'The big problem with silicon- based technology is that although it's mature and it works, it's too expensive. Materials and process wise, the dye-sensitised solar cell has the potential to be made much more cheaply than silicon.
'Ideally we'd want to do it for not a lot more cost than our current painted roofing, but especially in the beginning our main driver will be that it will be cheaper than the existing PV solutions.'
Wijdekop anticipates Corus will be able to produce a PV coating on steel in five years' time.
The project will benefit from input from Prof Michael Graetzel and Dr Brian O'Regan, a research fellow at Imperial, who collaborated on the invention of DSSCs, also known as Graetzel cells.