The April 2023 issue of IEEE Spectrum is here!

Close bar

Ultrathin Metasurface Displays Take Aim at the LCD

“Invisibility cloak” physics powers these super-high-res screen prototypes

5 min read
orange and yellow box on a gray background

The metasurfaces used as prototype display screens are thinner than liquid crystal cells, offer a greater resolution, and could consume less energy.

Andrey Miroshnichenko/UNSW Canberra

Currently, LCD screens are the most dominant and popular display technology for televisions and monitors, but they are unlikely to get significantly better in the future. Now a new study finds the kind of physics that make microscopic “invisibility cloaks” possible may lead to next-generation “metasurface” displays roughly 1/100 the thickness of the average human hair that could offer 10 times the resolution and consume half as much energy as LCD screens.

LCD technology depends on liquid crystal cells that are constantly lit by a backlight. Polarizers in front and behind the pixels filter light waves based on their polarity, or the direction in which they vibrate, and the liquid crystal cells can rotate the way these filters are oriented to switch light transmissions on and off.

LCD screens do continue to see advances by improving the liquid crystals, the display technology or the backlight. “However, improvement on LCD technologies are now mostly just incremental,” says Eric Virey, senior display analyst at market research firm Yole Intelligence in Lyon, France.

The prototype four-pixel device can switch its light transmittance with less than 5 volts in just 625 microseconds—which would translate to more than 1,000 frames per second.

One possibility scientists are exploring for next-generation flat-screen displays are metasurfaces, which are engineered to possess features not generally found in nature, such as the ability to bend light in unexpected ways. Research on metasurfaces and other metamaterials has led to invisibility cloaks that can hide objects from light, sound, heat, and other types of waves, among other discoveries.

Optical metamaterials, which are designed to manipulate light, contain structures with repeating patterns at scales that are smaller than the wavelengths of light they influence. However, their structures are typically static. This is an obstacle to many applications that require changeable optical properties, such as displays.

Previous research has investigated a number of different ways to electrically tune metasurface properties. However, until now, none of these approaches could simultaneously enable fast, large, transparent, solid-state adjustability, which are needed for use in displays and lidars.

In one new study, however, researchers experimented with electrically tunable metasurfaces compatible with standard CMOS production techniques. These rely on silicon’s large thermo-optical effect—that is, a change in temperature can significantly alter silicon’s optical properties.

“Our metasurface pixels are compatible with current silicon-chipmaking technologies, which keep the production costs low,” says study cosenior author Mohsen Rahmani, a professor of engineering at Nottingham Trent University, in England.

The core of the new fully solid-state device consists of a silicon metasurface—specifically, a 155-nanometer-thick film with 78-to-101-nm-wide holes arranged in a precise array within it. This metasurface is encapsulated by transparent electrically conductive 380-nanometer-thick indium tin oxide strips that can serve as electrically driven heaters.

“One of the important directions in the field of metasurfaces is the need for reconfigurability,” says electrical engineer Andrea Alù at the City University of New York Graduate Center, who did not take part in this research. This new work “enables a fast, efficient and compact way to tune the response of metasurfaces, which advances the field.”

“There is no need for significant investments in brand-new production lines to get this technology integrated.”
—Mohsen Rahmani, Nottingham Trent University

The prototype four-pixel device can switch how much visible and near-infrared light it transmits by ninefold with less than 5 volts in just 625 microseconds—which, without taking other factors into account at least, would translate to 1,600 frames per second. The technology, in other words, has a frame rate more than 10 times as fast as that of present-day video. The researchers detailed their findings online 22 February in the journal Light: Science & Applications.

“I find this work quite remarkable, especially the ability to dynamically modify the state of the metasurface, and do it at such high speed,” says Virey, who did not take participate in the new study. “It shows that there’s a lot of still unexplored potential applications and properties of metasurfaces and we’re probably just scratching the surface.”

How do metasurface displays compare to LCDs?

The scientists note a key advantage of this new approach is stability. “Silicon nanostructures are known for durability, which is one reason they are still the most popular material in the microchip industry,” says study cosenior author Dragomir Neshev, a professor of physics at Australian National University in Canberra. “We have run our prototype samples many times over a few months and did not experience any degradation.”

A common question the scientists receive about their work is the speed of cooling, “which is still on the order of human eye response,” Neshev says. He notes “one can significantly increase the cooling effect with further engineering, such as employing active cooling methods, or using air grooves around the pixels.”

The researchers say the new metasurfaces can replace the liquid crystal layer in LCD displays. At the same time, they would not require the polarizers that liquid crystals do, which are responsible for half of the wasted light intensity and energy use in these displays.

“In the longer term, there are also a lot of efforts on microLED technologies and electroluminescent quantum dots. So overall, the display industry is not running out of steam for innovation.”
—Eric Virey, Yole Intelligence

Current production lines for LCD displays can, with minimal modifications, be updated to replace liquid crystal pixels with metasurface ones, Rahmani notes. “There is no need for significant investments in brand-new production lines to get this technology integrated,” he says.

As promising as metasurface displays are, Virey cautions that organic LED (OLED) displays, which are currently the main rivals of LCDs, are strong competitors, and do not need liquid crystal layers.

“OLEDs are already used in about half of smartphones,” Virey says. “Adoption in TV is finally taking off, so is adoption in notebook. LCD won’t disappear and will likely remain the technology of choice for the bulk of entry-level to midrange displays, but their space is shrinking. As a result, while they’re planning on keeping their LCD manufacturing infrastructure, display makers have, for the most, stopped investing in new LCD fabs.”

Rahmani argues that OLEDs are “expensive and have a short life span,” Rahmani says. Silicon has a relatively long lifespan, the researchers note.

However, “OLEDs are progressing rapidly in term of cost, performance, and manufacturing processes,” Virey notes. “Samsung recently commercialized a new OLED architecture for TVs and monitors, combining blue OLED and quantum dots, a.k.a. ‘QD-OLED.’ More efficient blue OLED materials should hit the market in the next couple of years and help improve brightness, increase lifetime and reduce power consumption.”

In addition, “in the longer term, there are also a lot of efforts on microLED technologies and electroluminescent QDs,” Virey says. “So overall, the display industry is not running out of steam for innovation.”

In the hopes of breaking into the display market, the researchers now hope to optimize their device by tinkering with the heater dimensions, electrical input, and cooling approaches. Artificial intelligence and machine-learning techniques could also help design smaller, thinner, and more efficient metasurface displays, they add. Alù, of the City University of New York, notes smaller pixel sizes are also desirable.

The scientists aim to build a large-scale prototype that can generate images within the next five years. They hope to integrate their technology into flat screens available to the public within the next 10 years.

Such a timeline makes sense to Virey. “Looking back at OLEDs, electroluminescence of polymer materials was discovered in the ’50s, the first practical device was demonstrated more than 30 years later, and the first commercial displays didn’t hit the market until the early 2000s,” Virey says. “Research on microLED displays started in the early 2000s, and we’re not expecting to have the first high-volume commercial devices before 2025.”

Given how LCD manufacturers have spent more than a $100 billion on current fabs, “display makers might be happy to find a new technology that could give their aging LCD fabs a second life,” Virey says. “The researchers should make all efforts possible to ensure that it’s as compatible as possible with existing LCD manufacturing infrastructure. Can the same thin-film transistor process be used? Can you integrate the technology in an existing LCD fab with minimum change and investments?”

The Conversation (0)