Researchers Levitated a Small Tray Using Nothing but Light

One day a “magic carpet” based on this light-induced flow technology could carry climate sensors high in the atmosphere—wind permitting.

In the basement of a University of Pennsylvania engineering building, Mohsen Azadi and his labmates huddled around a set of blinding LEDs set beneath an acrylic vacuum chamber. They stared at the lights, their cameras, and what they hoped would soon be some action from the two tiny plastic plates sitting inside the enclosure. “We didn’t know what we were expecting to see,” says Azadi, a mechanical engineering PhD candidate. “But we hoped to see  something.

Let’s put it this way: They wanted to see if those plates would levitate, lofted solely by the power of light. Light-induced flow, or photopheresis, isn’t a breakthrough on its own. Researchers have used this physical phenomenon to float invisible aerosols and sort particles in microfluidic devices. But they have never before moved an object big enough to grasp—much less lifted anything that can carry objects itself.

And it worked. “When the two samples lifted,” Azadi says, “there was this gasp between all four of us.” The Mylar plates, each as wide as a pencil’s diameter, hovered thanks to nothing but the energy from the light below, according to a paper published today in Science Advances. Energy from the LEDs heats up the Mylar’s specially-coated underbelly, energizing air particles under the plastic and propelling the plates away with a tiny, but mighty, gust.

This engineered structure is the first instance of stable photophoretic flight, and Azadi’s accompanying theoretical model can simulate how different flying plates would behave in the atmosphere. In particular, the model indicates that a levitating plate could mosey 50 miles overhead while carrying sensor-sized cargo. It’s an idea the lab members have floated as a way to study weather and climate—although atmospheric scientists say the idea is still preliminary and will face some daunting meteorological challenges.

Courtesy of Mohsen Azadi

There’s a reason why scientists would want to get a tiny sensor into the under-explored mesosphere, which lies between 31 and 53 miles above your head. “Sometimes it’s called ignorosphere, in joke,” says Igor Bargatin, a mechanical engineering professor at Penn and Azadi’s adviser, who led the study. “We just don’t have access to it. You can send a rocket for a few minutes at a time, but that’s very different from doing measurements using airplanes or balloons.”

We haven’t ignored the mesosphere because it’s uninteresting; we’ve ignored it because it’s out of reach. The denser air below it affords enough lift to planes and balloons. And the thermosphere above is thin enough that air drag doesn’t burn orbiting satellites. The mesosphere gets the worst of both worlds—it’s too thin for lift but thick enough to burn an orbiter.

That’s a drag for scientists, because the mesosphere is loaded with interesting phenomena, like weird blue and red lightning and the microscopic shrapnel of millions of meteors—shooting stars—scorching through it every day. The chemistry in that layer is also valuable for scientists interested in tracking ozone damage, according to Daniel Marsh, an atmospheric scientist at the National Center for Atmospheric Research. “Solar storms cause energetic particles to enter the mesosphere, creating nitric oxide,” Marsh wrote in an email to WIRED. That nitric oxide seeps lower into the atmosphere and eats away at Earth’s protective stratospheric ozone.

Sending scientific feelers directly into this zone requires engineering a whole new way of flying, Bargatin says. And using light makes sense because of its intrinsic energy. Scientists have tested the idea of catching light particles’ momentum in solar sails to travel into deep space at 10 percent of lightspeed, but that idea collapses in the mesosphere’s gravity. Over the last century, physicists have gotten more comfortable using light to move matter in other ways. Lasers can nudge proteins and beads, sort cells, and pluck molecules like tweezers, for instance. “Pretty much all the research that has been done so far focused on microscopic particles,” says Bargatin. His lab published a paper in Advanced Materials last year reporting a hollow aluminum-based sheet that could hover over an air cushion. But this new study comes with higher hopes—designing a flight system so stable that researchers could simply let these devices loose in the mesosphere.Courtesy of Mohsen Azadi

Azadi began with the basics, diagraming levitator designs and charting which physical forces might cause light to propel a surface. He ran thought experiments as simple as imagining throwing spheres against a wall. “What can we do to the surface of the wall so when we throw a sphere at a wall and it bounces back, it bounces back faster?” says Azadi.

“I would just have a piece of paper and a pen, and try to sketch different things,” he continues, “and make those very simple thought experiments into mathematical, rigorous formulas.”

The team eventually landed on a design: a flat disk with two distinct faces. For the top, they chose Mylar, that shiny plastic used in thermal blankets. Mylar is cheap, light, and smooth, and some versions are unfathomably thin—only 500 nanometers thick in this case. That’s 50 times thinner than household ClingWrap, and so slim that it’s actually transparent. For the underside, Bargatin’s team coated the Mylar surface with a shag carpet of tiny rod-shaped carbon threads called carbon nanotubes. Each nanotube is only a few atoms across and about as long as a strand of hair is wide.

After an ambient gas molecule from the air collides with a warm object, it picks up a small amount of energy and bounces off faster than it arrived. (Thermodynamics dictates that a hotter particle is a faster particle.) But not every surface transfers that energy to gases equally. Some, like a smooth sheet of Mylar, spring gas molecules away with only a little boost. Other surfaces, like a tangled mess of carbon nanotubes, can trap and heat gas molecules so much that they fire away a lot faster.

When this jet-black carbon carpet absorbs light, its tangled mess of nanotubes warms. Gas molecules that slip into the shag then collide with so many nooks that they heat up more than the molecules ricocheting off the smooth upper surface. This rush of molecules shooting down from the bottom surface faster than up from the top creates a lift force, says Bargatin. “You throw enough molecules down, you’re gonna create a jet,” says Bargatin. “That’s what helicopters do.”

On that day in late 2019 when Azadi and the rest of the team gathered around the vacuum chamber to try out the nanotube design for the first time, Azadi let the mini magic carpets float a few millimeters above the surface at mesosphere-like pressure. In one instance, two mylar plates circled each other as though they were dancing. “We decided to name the move because it worked so beautifully,” Azadi says. “It looked like two of them danced with the same very harmonic dance. It was like, let’s call it ‘Tango.’”

By surrounding one central LED with a ring of more intense LEDs set beneath the vacuum chamber, they were also able to demonstrate stable levitation. This setup keeps the levitating plate confined to an optical trap—if the plate begins tilting and zooming away, the light boundary forces it back to the center. Levitating without this balancing force is like balancing a pea on the underside of a spoon.

“When they said that they have a centimeter-sized object that they can levitate using photophoretic forces, I was very skeptical,” says Yael Roichman, a physicist at Tel Aviv University who was not involved in the study. Roichman studies optical trapping and has used lasers to levitate dust particles. Conventional photopheresis experiments rely on a temperature gradient—a hot face and cold face—to propel objects. This restricts an object to only moving away from an energy source, nixing hopes of sun-powered levitation. But she says Bargatin’s idea is different. Regardless of where the light originates in relation to the levitator, it will reach the down-facing nanotubes and provide lift. “What they did doesn’t depend on the temperature gradient, which gives you very small forces, but depends on something completely different,” she says. “I think this is actually potentially very useful and innovative. It seems simple, but it’s not simple.”Courtesy of Mohsen AzadiCourtesy of Mohsen-Azadi

Immediately after Azadi first captured the levitation, he rushed to his computer and punched the experiment’s exact physical parameters into his theoretical model. The hovering behavior they observed matched the theory they had developed. “The range of pressure that it works at, the range of light intensity where the forces maximize—they all matched what I had seen,” says Azadi. “So that was a very exciting moment, to see that the theory works and it matches the experiments really well.” That validation meant they could now use their model to predict how microflyers of different sizes would behave in any atmospheric condition. They could calculate, for instance, the diameter of a plate that could carry the heaviest payload at a particular altitude without being too wide to float.

Their simulations estimated that a 6-centimeter plate could carry 10 milligrams of cargo in the mesosphere under natural sunlight. Ten milligrams may not sound like much; a drop of water weighs five times as much. But engineering advances have shrunk silicon chips into dust-sized sensors far smaller than that. These “smart dust” systems can fit a power source, radio communication, and a data-collecting sensor in cubes only a millimeter across. “Researchers can do a lot when you give them a cubic millimeter of silicon,” says Bargatin. “And a cubic millimeter of silicon weighs a couple of milligrams.”

In their vacuum chamber test, they found that when cranking the light intensity up past the power of sunlight, that extra rush of energy carried the flyer higher. But after about 30 seconds, the disk began curling up from photophoretic force, eventually collapsing. Ultrathin Mylar is very flimsy on its own, says Bargatin. The shag of carbon nanotubes makes the Mylar disk more rigid, but the force of high-speed molecular collisions eventually buckles the flyer. The team’s model can predict what disk sizes, air pressures, and light intensities cause this, and Bargatin says work to develop a lightweight frame is ongoing.

Bargatin envisions researchers one day releasing sensor-laden levitators in the mesosphere and letting them roam, like weather balloons or floating ocean sensors. “Another approach is to actually develop smart flyers that can control where they’re going,” he says. The same tilting that stabilizes the levitators could be used to steer them. And, he adds, suspending the sensor from the levitator like a parachuter hanging from a canopy would help keep the system upright when faced with wind.

Still, Marsh is not convinced that such a device could withstand mesospheric conditions. “Any instrument is going to have to operate in the extreme conditions of the mesosphere, where the average winds can easily exceed 100 mph,” he writes. Winds in the upper mesosphere can be especially shearing, temperatures can drop to 140 below zero, and space weather radiates through the mesosphere and can damage communication systems.

Paul Newman, chief scientist of Earth Sciences at NASA’s Goddard Space Flight Center, agrees that accounting for mesospheric wind will be a big technical challenge, but he can’t help but delight at the possible applications. “I actually think this is a really cool idea,” he says. One possibility would be to probe water vapor in the mesosphere, where polar clouds form so high that the sun still illuminates them at night. The mysterious clouds aren’t just beautiful, Newman says; their possible link to increased greenhouse gases means they may become more common—but researchers can’t track the mesosphere’s water content and temperature as well as they’d like. Mesospheric clouds are “another sign of climate change. And we need information to show that,” Newman says. “That’s why these could be really cool for getting data on atmospheric composition.”

Newman adds that the plates’ tininess and levitation ability could also be intriguing for Mars research. The air pressure of the Martian atmosphere is similar to Earth’s mesosphere, so perhaps light, autonomous levitators could collect temperature or composition measurements. “You can just take off once per day, and go up and then come back down and land on your little Martian lander,” he imagines. “We don’t have that information on Mars. That would just be fantastic.” (NASA is planning to test out a small helicopter called Ingenuity as part of its soon-to-land Perseverance rover mission, but that craft will be much bigger and is still in the test flight stage; it’s not ready for science missions yet.)

Bargatin says they are currently exploring applications for Mars, and that the team is also hoping to make their microflyers work at sea level on Earth. But regardless of any eventual use, Azadi will always remember seeing the Mylar creation float for the first time, exactly according to his theoretical predictions. “After that,” he says, “I called my girlfriend and I said, ‘I think I’m going to graduate soon.’”

All Rights Reserved for Max G. Levy

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.