Skip to Content

Nanopatterns Improve Thin-Film Solar Cells

Thin-film silicon solar cells are more efficient with tiny holes in the back electrical contact.
October 19, 2009

Any given solar-cell technology has drawbacks and advantages. Thin-film solar cells, for instance, require less material than traditional solar cells, and are therefore cheaper, lighter, and flexible. And if those thin films are made with amorphous silicon, the cost is further reduced. The problem, however, is that thin-film solar cells made of amorphous silicon tend to have extremely low efficiencies compared to thicker, crystalline silicon photovoltaics.

Rainbow response: The absorption of light with a wavelength of 660 nanometers is greatest for the spacing and diameter of the holes in the dark-red portion of this graphic. The inset shows the layout of the cell.

But now, research from Caltech shows that it’s possible to increase the efficiency of thin-film amorphous silicon cells 37 percent–from 4.5 percent efficiency to 6.5 percent, which is still significantly lower than commercial crystalline silicon cells that achieve efficiencies of more than 30 percent–by simply adding a pattern of nanoscale holes to the electrical contact on the back side of cells. Importantly, the research, led by Harry Atwater, professor of applied physics and materials science at Caltech, appears to be practical for scaling up to large-scale production of the cells.

A number of researchers and startups are exploring thin-film solar cells made of nonsilicon materials. But, says Atwater, these materials are relatively rare, and as such, they aren’t practical for widespread use. “These represent challenges at extremely large scale,” he says.

Silicon has the great advantage of being abundant and of having a long history in the manufacturing of electronics. But as a thin solar cell, silicon is less than ideal. There is a mismatch between the distance it takes for photons to be absorbed in silicon and the distance traveled by electrons that produce electrical current. Essentially, electrons knocked out of position by photons tend to spring back to their spots before they can be collected, resulting in low efficiencies in converting sunlight to electricity. But, if the optical absorption can be improved, then more electrons overall can be collected, increasing efficiencies.

Researchers and companies are exploring various options for improving efficiencies in such cells. For example, StarSolar, an MIT spinoff in Cambridge, MA, is exploring photonic crystals, structures that reflect light many times within the solar cell to increase the chance it has to produce electrical current. But so far, the approach appears to be difficult to scale up.

Atwater’s approach targets the back side of the solar cell, the metal electrical contact that sits below layers of “active” silicon material where photons are absorbed. Instead of using a grating to produce multiple internal reflections, he’s using an array of holes 225 nanometers in diameter. When light hits metal with an array of holes at this scale, some interesting physics occurs. The energy from the light is essentially trapped onto the two-dimensional wave on the surface of the metal. The electrons in these surface waves, called plasmons, are easier to use for generating an electrical current than those in silicon, which quickly snap back into place.

In previous work, Atwater and others have explored plasmonics to improve efficiency in cells made of gallium arsenide, a semiconductor material commonly used in optics; they have also tried organic materials and even amorphous silicon. However, this previous work applied the plasmonic structures to the front of thin-film amorphous silicon solar cells; in this position they tend to absorb light of certain wavelengths and convert it to heat.

“One of the things that one sees with particles on front side of solar cells is a net loss of short wavelengths due to resonant absorption in the metal itself,” says Atwater. “We want to avoid that.”

“It was very clever to put the metallic structures in the back contacts,” says Mark Brongersma, professor of materials science and engineering at Stanford. “Now, all the incident light gets at least one pass through the cell, and the plasmonic structures can be optimized to manage a few photons with a narrower spectrum.” In other words, the size and spacing of the holes can be tweaked to take advantage of the wavelengths of light that make it through the silicon to the back contact.

To make the nanopatterned holes, Atwater’s group uses a stamp that is able to imprint holes over the area of an entire silicon wafer. In a series of simple steps, the array of holes is formed in a thin layer of material on the silicon wafer, which is subsequently covered with metal. The active silicon material in the cell and the top electrical contact are then deposited on top of the patterned back side. The stamp can be used for thousands of imprints before it needs to be replaced.

Brongersma, who was not involved in the work, adds that this fabrication technique is certainly amenable to high-volume manufacturing. “The work also makes a big step forward by showing that we may be able to scale plasmonic photovoltaic up to large areas.”

The researchers, who will publish the work in an upcoming issue of the journal Applied Physics Letters, have run simulations to determine the optimum hole diameter and spacing. To test the performance of different types of holes, the researchers focused on the efficiency of a single wavelength of light, 660 nanometers. Their simulations indicate that by increasing the diameter and reducing the depths slightly, they can improve absorption at that wavelength from 42 percent to 54 percent, which should further improve the overall performance of the entire photovoltaic, although it’s unclear by how much.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.