Research removes obstacle to lasers and computers working at speed of light
CLEMSON, South Carolina — By juxtaposing contrasting properties of two different nanomaterials, a team of Clemson University physicists has developed an optical diode that enables light to move in one direction, a step that could lead to high-energy lasers and optical computers capable of processing data at the speed of light.
Presently, computers use the movement of electrons to process data. Through electrical components like transistors and diodes, the flow of electrons can be turned on and off or made to flow exclusively in a certain direction like a one-way street. Without these devices, computing as we know it would not exist.
The drawback to electronic computing, however, is that the speed at which computers can perform calculations and process information is limited by how quickly electrons can navigate within solid devices like diodes and transistors.
Ramakrishna Podila, a professor in Clemson’s department of physics and astronomy, equates the phenomenon to people trying to move about in a dense crowd.
“Crudely speaking, excited electrons are like being in a crowded mall looking for your friend. As you go toward your friend, you’re going to bump into a lot of other people and also get pushed into walls and doors,” Podila said. “In a solid, these walls and doors are the motions of atomic nuclei called phonons and the other people are the electrons. So when an electron is going through a solid, it will bump into other electrons and phonons. That will slow it down, for one, but it could actually make reverse progress, too, because when the electron bumps into something, it falls back.”
To evade the limited mobility of electrons, many physicists are in pursuit of an optical computer that can function using photons, or particles of light, rather than electrons. Traveling at a speed of 200,000 miles per second, photons are the fastest thing in the universe and could literally boost computing to lightning speeds. However, this idea comes with its own set of limitations.
“The problem is, unlike electrons, you cannot make photons follow one direction; meaning, if you see me, I see you, too. This is called time reversal symmetry,” Podila said. “If I have a laser beam, like a laser pointer, and I put my hand in front of it, I can see the laser. But if I hold it in my other hand and put my opposite hand in front of it, I will still be able to see the laser.”
Making light flow in just one direction is critical for many applications, such as high-energy lasers and optical computing, he said. Scientists have achieved this feat before in devices such as the Faraday rotator, which exploits a magnetic field to rotate the direction of polarized light being outputted. But the Faraday rotator is bulky and requires a magnetized element to be successful, which places a limitation on how it can be used.
Five years ago, Podila and colleagues at the Clemson Nanomaterials Institute had the idea to make an optical diode that has everything the Faraday rotator doesn’t — one that is tiny, scalable and does not require magnetism or polarization to function. Modeling the optical diode on the concept of an electrical diode, the team planned for a device that could work with any kind of light at any intensity.
It works by placing two different nanomaterials with different optical properties side by side, one a saturable absorber and the other a reverse saturable absorber.
“With very weak light, the saturable absorber has low transmission; some light makes it through. But if you have really intense light, it becomes more transparent. The brighter the beam is, the more transparent the material is. A reverse saturable absorber is the other way around: the brighter the beam is, the more opaque the material becomes,” Podila said. “When light goes through a saturable absorber, then the reverse saturable absorber, it makes it out. But when it flows the other way on a reverse saturable absorber, it doesn’t make it through because these are nonlinear effects.”
Initially, the team chose fullerene, a 60-carbon molecule, to be the reverse saturable absorber and graphene, a single layer of carbon atoms, to be the saturable absorber. Since then, the idea has become a precursor to other pioneering works on optical diodes that have received attention in recent times.
“The problem with graphene is that it’s one atom thick and it’s all carbon. So while it is great for many things, when you have a very strong, bright beam of light, the graphene burns due to oxidation,” said Yongchang Dong, a graduate researcher and the first author on this study. “We proved the concept of an optical diode, but we needed a material that could withstand higher energies.”
Thus, the team went to work, using hydrofluoric acid to selectively pull aluminum atoms out of a titanium-aluminum-carbon mixture. The resulting material – two-dimensional titanium carbide – can withstand the high energies of light and be manufactured into different thicknesses, adding durability. Podila’s collaborators – Yury Gogotsi and Vadym Mochalin at Drexel University and Missouri Science & Technology, respectively – named the material MXene after its graphene-like morphology.
At the moment, the team’s development has applications in high-powered lasers, allowing for pulse shaping that can increase the intensity of a laser pulse over a shorter period of time.
“If I have one laser pulse that is spread out over time and another one that spikes, overall the total energy might be the same between the two, but the first one doesn’t pack much power,” Dong said. “It’s like retirement savings. If I’m putting $100 in every day, I cannot buy an airplane right now, but 50 years from now I’ll have $5 million, and then I’ll be able to. If I could pack the $5 million into one day, then I’d have so much power. That is pulse shaping, and that is how you can get very high-powered lasers.”
Optical computing, Podila said, is still a long way away. The next step in its creation would combine an optical diode with a device called a logic gate that can perform binary input and output operations.
The work outlined in this study was published in January 2018 in Advanced Materials. Research reported in this publication was supported by the National Science Foundation (grant number DMR-1310245), Missouri University of Science & Technology and Clemson University. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies.