One of the challenges facing designers of traditional flat solar panels is the fact that the sun doesn't conveniently stay in one place. This means that in order for a panel to receive as much sunlight as possible, it has to pan with the sun as it moves across the sky. While there are motorized assemblies designed to do just that, they add complexity, weight and expense to photovoltaic systems. Now, however, University of Michigan scientists have developed a simpler alternative – and it's based on the ancient Japanese cut-paper art of kirigami.
The U Michigan engineers consulted with Matthew Shlian, a paper artist who also lectures at the university's School of Art and Design. He showed them a kirigami pattern that would suite their purposes, which basically consisted of stacked lines of dashes cut into a piece of paper.
Doctoral student Aaron Lamoureux and associate professor Max Shtein reproduced an advanced version of that pattern on a sheet of Kapton plastic, that already had individual solar cells adhered to it.
When left alone, that sheet sits flat. When it's stretched, however, the strips of plastic between the cuts (and the solar cells that are on them) twist to one side – it's possible to finely control how much they twist, by modulating the degree to which the sheet is stretched. Mounted under glass in a flat photovoltaic panel, the cells can turn to stay facing the sun, even though the panel itself doesn't.
When tested in a setup simulating the summer solstice in Arizona, it was found that the kirigami panel was able to produce 36 percent more energy than a traditional panel. A conventional motorized sun-tracking system only performed slightly better, at 40 percent.
"We think it has significant potential, and we're actively pursuing realistic applications," says Shtein. "It could ultimately reduce the cost of solar electricity."
A paper on the research was recently published in journal Nature Communications. The sheet's twisting action is demonstrated in the following video.
Source: University of Michigan