Skip to Content

Expandable Silicon

A new chip design could lead to cheaper solar panels, sensor networks, and flat-screen TVs.
December 14, 2007

A new design for silicon-based chips makes it possible to mechanically stretch them out to cover large areas. These expanded chips, which could be thousands of times the size of the original, could be used to make cheaper solar panels, sensor networks, and flat-screen TVs.

Growing chips: Researchers have built chips (top) that can be expanded for large-area applications (bottom).

The chips, built by researchers at Stanford University, consist of free-floating islands of silicon surrounded by coils of silicon wire. Each island can be processed to include transistors, sensors, or materials for tiny solar cells. When the corners of the chip are pulled on, the coils around the silicon islands unwind. As they do, the islands, which start out nearly touching each other, spread apart. The end result is a netlike array of silicon devices.

So far the researchers have demonstrated arrays that are 50 times larger than the original chip, but they’ve been limited by the size of their laboratory equipment. Peter Peumans, the professor of electrical engineering at Stanford who led the work, says that the chips could be made to expand thousands, or even tens of thousands, of times. Peumans’s work was presented this week at the International Electron Devices meeting in Washington, DC.

Silicon nets: Using conventional silicon-processing techniques, researchers at Stanford University have built chips that consist of islands of silicon surrounded by silicon coils. The top-left image shows one such silicon island, and the bottom-left image shows the entire chip composed of an array of these islands. When the corners of the chip are pulled, the coils unwind, and the islands spread apart. The finished network is shown at the bottom right. The top-right image shows the coils completely unwound.

The work is “taking the integrated circuit concept that has been so successful in microelectronics and adapting it to large-area applications,” says Marc Baldo, a professor of electrical engineering at MIT. The semiconductor industry has excelled at packing more high-performance transistors into a given space, driving down the cost per transistor in the process. But many applications require that transistors and other silicon-based devices be more distributed.

For example, flat-screen TVs need millions of transistors spread out to control each pixel. For LCD TVs, it’s been possible to use relatively low-performance transistors, which can be made by depositing amorphous silicon onto large pieces of glass. But the next generation of brighter, more colorful, and more energy-efficient displays, such as organic LED displays, require much higher performance transistors created from higher-grade silicon, which can be extremely expensive, making it impractical to coat an entire display with it. With Peumans’s method, it could be possible to use only a small amount of high-grade silicon, thereby cutting costs. What’s more, the devices are already wired together. That’s an important advantage over some other methods for making large-area electronics since “wiring up large-area electronics can be very expensive,” Baldo says.

The ability to use less silicon, and to form orderly arrays of prewired silicon devices, could also be useful for making cheaper solar panels. In conventional solar panels, light is absorbed because the entire panel is coated with high-grade silicon. Now a number of companies are reducing the amount of silicon needed by concentrating sunlight onto smaller silicon chips. For example, one company makes an array of small lenses that focus light onto even smaller silicon solar cells. Peumans says that his method offers a cheaper way of making such solar-cell arrays. Earlier this year, he founded a company called NetCrystal, based in Mountainview, CA, to make such panels, which he expects can be created for a third of the cost of today’s panels.

Peumans is also working with Boeing to develop sensor networks for airplanes. The goal is to distribute high-performance, silicon-based sensors between layers of the composite materials that make up the wings and other parts of new aircraft, such as the Boeing 787. These sensors would be used to determine if the materials are cracking or delaminating. The sensors could decrease downtime for inspections and help maintenance crew spot problems earlier, Peumans says.

Peumans’s technology is not the first attempt to make large-area electronics. Other approaches, however, tend to produce devices that fall considerably short of the performance of chip-grade, single-crystalline silicon. Some researchers, for example, are developing inexpensive methods that use commercial printing techniques to deposit inorganic or organic semiconductor “inks.” But the best inorganic ink-based devices perform about an order of magnitude worse than single-crystalline silicon, whereas organic ink-based transistors are a thousand times worse.

The biggest hurdle in developing Peumans’s approach was showing that the coils around the silicon islands would be strong enough not to break as they unwind, but he demonstrated a way to treat the coils to make them stronger. The next step is to demonstrate functioning devices. He has already developed prototype solar cells and is working on partnerships to develop other applications.

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.