The most efficient solar cell ever made can convert almost half of the sunlight that reaches it into electricity. If widely available, such a powerful cell – with double the efficiency of the standard solar panel on sale today – could slash the cost of solar energy, as well as the amount of land and material needed to build it.
 |
| An Oxford PV solar module with layers of perovskite and silicon Oxford PV |
The problem is that such ultra-efficient cells were designed to power satellites in space and were made using extremely expensive materials. Those dollars are a drop in the bucket for big budget projects in orbit. But it rules out using those designs to generate electricity here on Earth, where cost largely decides what types of energy systems get created.
But researchers are now beginning to make that space-age solar power viable on the ground, using similar design principles to combine multiple layers of sunlight absorbers, but with lower-cost materials and easy-to-manufacture approaches.
The result could be cheaper and more versatile solar power, and a faster transition away from fossil fuels.
Squeezing silicon
Solar cells use a semiconducting material called an absorber to convert the sun’s radiation into an electrical current. The first practical photovoltaic (PV) cell, designed at Bell Labs in New York City in the 1950s, used a silicon absorber that could convert about 4 per cent of the solar radiation that reached it into electricity.
Since then, solar cells have become far more efficient at converting sunlight to power. Today, the typical solar panel – the type that dominates the market – has an efficiency of around 20 per cent. Along with advances in manufacturing, that has helped make solar power the cheapest and fastest-growing form of energy in much of the world.
Even more efficient cells would further accelerate the solar boom by cutting costs, as well as reduce the large amounts of land and materials required, says Sam Stranks at the University of Cambridge. This is especially true for rooftop solar and other uses constrained by space, where small efficiency gains can have outsized effects on cost.
There are ways besides efficiency to increase the energy yield from solar installations, such as by keeping panels clean, keeping them cool by floating them on water or growing vegetation underneath. But silicon cells themselves can’t get much better: the best silicon cells in lab settings have reached efficiencies over 27 per cent, which isn’t far from their theoretical maximum of around 30 per cent.
“There’s not much more to squeeze out of silicon,” says Zachary Holman at Arizona State University.
That limit exists because silicon, like other materials, can only absorb certain wavelengths of light – in silicon’s case, that is the infrared part of the spectrum. Other materials absorb different bands of radiation, but all hit a ceiling at just over 30 per cent. (For a true physics lesson on this, read up on the Shockley–Queisser limit.)
The challenge now is to develop practical and affordable designs that can break that 30 per cent efficiency limit, says Nancy Haegel at the National Renewable Energy Laboratory in Colorado. “It’s kind of a holy grail for this community.”
Adding layers
The main approach to surpassing silicon is to stack different materials together within a cell, each layer absorbing different wavelengths of radiation. Cells with two layers are called tandem cells, while multi-junction cells combine more than two layers. The maximum theoretical efficiency for a multi-junction cell with infinite layers under concentrated sunlight is a whopping 86.8 per cent.
Infinite layers are impossible, of course, but actual cells have performed quite well. In the lab, both tandem and multi-junction cells have achieved efficiencies above 30 per cent. The current record of 47.6 per cent was set in 2022 using a cell with four layers, although that feat required radiation concentrated by lenses to the intensity of 665 suns. The record under the radiation of a single sun is just under 40 per cent.
Practical versions of these super-efficient cells have been used to power orbiting satellites and space stations for years, including multi-junction cells with as many as six layers. But the exotic materials and advanced processes required to build them mean those space cells are thousands of times more expensive than regular silicon cells.
“The question for terrestrial solar has been: how do we do what they’ve been doing for space, but with materials that are dirt cheap?” says Holman.
Perovskite on silicon
The first of the space-inspired solar cells to be available on Earth looks set to be designs that use a silicon cell topped with a layer of perovskite – a crystal made up of titanium, calcium and oxygen. This versatile class of materials can absorb visible light to complement the infrared absorbed by the silicon, raising the theoretical of such perovskite-on-silicon tandem cells to 43 per cent. Perovskites are also more flexible, lighter and less energy-intensive to make than silicon crystals.
Research versions of these cells have set a stream of records and commercial versions have been just around the corner for several years. But they now appear to be ready for launch.
Within the next two months, a UK company called Oxford PV – widely considered to be a front-runner in the field – will ship panels for its first commercial installation of perovskite-on-silicon cells, Chris Case, the company’s chief technology officer, told me.
The panels, produced at the company’s factory in Germany, will be installed as part of a 20-megawatt solar facility owned by a large utility in the US, providing between 100 kilowatts and 1 megawatt of power capacity, using cells with greater than 27 per cent efficiency, says Case.
Other perovskite-on-silicon cells are even more efficient. Last month, a Chinese solar cell company called Longi set a new record of 34.6 per cent for a perovskite-on-silicon cell, breaking its own record from late last year. But Oxford PV is more advanced in bringing its technology to market, says Case. Last month, the company said a module made of up 60 cells designed for rooftops achieved 26.9 per cent efficiency, a record for a module of that size, which is closer to a practical panel than the individuals cells tested in labs.
Case wouldn’t provide more details on the US installation, but he says the company aims to manufacture gigawatts-worth of the cells within the next few years. “We need to build much bigger factories,” he says.
Holman says the company’s advances are “impressive” and the prospect of commercial perovskite panels is exciting. But he points out that even a gigawatt-scale plant would be a sliver of overall PV manufacturing, which has now reached around a terawatt – or 1000 gigawatts – of annual capacity.
To speed things up, Holman is involved with an Arizona-based company called Beyond Silicon working on making tandem cells using processes that can easily be added to existing PV manufacturing lines. Researchers are also developing other ways of producing large volumes of the cells, such as by depositing liquid perovskite onto the surface of silicon under a vacuum.
Proving such panels are reliable is another big challenge, says Haegel. Fragile perovskites degrade in sun, air and heat, and it isn’t clear how long the panels can maintain their high efficiencies through years of exposure to the elements – a requirement for solar developers building facilities they intend to last for decades.
“Can we guarantee our performance? Not really,” says Case. He says Oxford PV’s panels that have been tested outside for several years haven’t shown concerning evidence of degradation, and accelerated ageing tests have demonstrated degradation rates similar to those of silicon panels. However, those tests were designed for silicon alone, and there is disagreement over whether they are relevant to perovskite.
Printing panels
But assuming those issues are manageable, it could be just a few years before perovskite-on-silicon is a mainstream technology, says Stranks, who co-founded another perovskite start-up called Swift Solar. By 2030, perovskite tandems may have begun “a rapid takeover of the whole solar industry,” according a new report from Rethink Energy, a UK think tank. And that could be just the beginning. “Perovskite-on-silicon is a great stepping stone,” says Stranks.
Researchers are already exploring perovskite designs that do away with silicon entirely, using one, two or even three layers of different perovskites that each absorb different wavelengths of light. Such flexible, lightweight cells would not only be more efficient than silicon designs, but could be manufactured roll-to-roll like newsprint, rolled out on roofs like tarpaulin or added to windows, says Stranks.
And perovskites aren’t the only materials that researchers are exploring. “Perhaps as we learn more about these perovskite materials, we’ll find the next material,” says Haegel, enabling more of what has succeeded in space to be put to work on Earth. “One has reason to be very optimistic about where this will go.”
0 Comments