A new type of solar cell has broken a theoretical limit on the efficiency of silicon-based cells, which could allow us to get more power from sunlight.
Almost all commercial solar cells are made of silicon. These can only convert a narrow frequency band of sunlight into electricity. Light that is too far outside this range either passes through directly or is lost as heat, giving silicon cells a theoretical efficiency limit of around 29.4 percent.
In theory, this limit could be higher if another material that generates electricity from light in a different frequency range is stacked on top of the silicon layer. Perovskite, a crystal of titanium and calcium, is well suited to this because it absorbs light closer to the infrared spectrum better, but making it efficient has proven difficult. This is due to stray electrons being reabsorbed into the crystal before they can be converted into current.
Now two research groups have found ways to pair perovskite with silicon and achieve higher efficiency.
To make silicon and perovskite work together, Chin Xin-Yu at the Swiss Federal Institute of Technology in Lausanne and colleagues used a two-step process. The silicon cell is first coated with a tight layer of precursor chemicals, before a second layer of chemicals is added to convert the precursors to perovskite. This process causes fewer defects at the silicon-perovskite interface, Chin says, and therefore increases the number of electrons available for current. The team’s device has an efficiency of 31.2 percent.
In a separate study, Silvia Mariotti at Helmholtz-Zentrum Berlin and colleagues injected liquid piperazinium iodide into the perovskite shell, which also appeared to reduce stray electrons, achieving an efficiency of 32.5 percent.
“The efficiency is phenomenal,” he says kyle frohna at Cambridge University. However, these numbers are currently limited to solar cell sizes much smaller than would be required for commercial use, he says.
In May, solar firm Oxford PV demonstrated that perovskite-silicon tandem cells they could be made on a production-ready scale, although they had a slightly lower efficiency level of 28 percent.
“If we can do them at scale, which some companies seem to be able to do, great,” says Frohna. “The only caveat is that we want to make sure they are stable enough to last for a long period of time.”
