The playbook for perfect polaritons

(Press-News.org) Light is fast, but travels in long wavelengths and interacts weakly with itself. The particles that make up matter are tiny and interact strongly with each other, but move slowly. Together, the two can combine into a hybrid quasiparticle called a polariton that is part light,  part matter. In a new paper published today in Chem, a team of Columbia chemists has identified how to combine matter and light to get the best of both worlds: polaritons with strong interactions and fast, wavelike flow. These distinctive behaviors can be used to power optical computers and other light-based quantum devices.

“We’ve written a playbook for the ‘perfect’ polariton that will guide our research, and, we hope, that of the entire field working on strong light-matter interactions,” said Milan Delor, associate professor of chemistry at Columbia.

Delor’s lab is particularly interested in what are known as exciton-polaritons, which form when photons, the particles that make up light, combine with the energy from excited electrons in a material. They are particularly promising for creating light-based, high-speed optical computers.

To operate, computers rely on a system of tiny gates that open and close to regulate how information is transmitted. In today’s electronic computers, these gates are the transistors, which open and close as electrons interact with each other. Light could offer  a faster and more efficient way to transmit information, if photons interacted. But they don’t: Any beam of light will simply pass through another, making it a challenge to create computer gates with light alone.

Polaritons offer a path forward to power entirely light-based computers. Polaritons form when light interacts strongly with a material at the quantum level. Photons help excitations, like the energy from excited electrons that Delor’s group is interested in, become coherent, or synchronized, across large spatial scales—akin to how groups of fireflies will come to flash in unison. Disturb any one, and the effect will ripple through the larger whole. With polaritons, this wavelike propagation is a manifestation of long-range coherence and a potential means to transmit information at the speed of light (or pretty close to it), and at nanoscopic scales.

In 2023, Delor and his team developed an ultrafast imaging technique that can capture exciton-polaritons in motion. They saw waves—a sign of coherence—but only up to a point. As the hybrid particles became more “matter-like,” they increased their interactions but lost coherence due to an increased sensitivity to noise, such as disorder in the material.

“You can see the problem: when you combine light and matter, you don’t just inherit the best parts, you also inherit the worst,” said Delor. “Our game was to find systems that optimize the coherence that comes from light with the strong interactions from matter, while minimizing weaknesses.”

Delor, along with postdoctoral researcher Yongseok Hong and PhD student Ding Xu, GSAS ‘25,  began designing and testing materials in search of the best polariton-producing properties. Their samples ranged from films with randomly arranged molecules to more organized molecular crystals to the fixed lattices of different 2D materials.

They identified three guiding rules: the chosen material should have large optical absorption (a measure of how strongly light can interact with matter), low disorder (a reflection of the number of defects or impurities in the material), and a little bit of inherent exciton delocalization (how large the exciton radius is, before interacting with light). The latter in particular was an overlooked property that turned out to be the key missing ingredient, as some exciton delocalization ultimately protects polaritons against noise. Together, these properties help preserve polariton coherence, even in the presence of strong polaritonic interactions and inevitable disorder.

“This is exactly what’s needed for polaritons to realize their much-touted ability to ‘combine the best of light and matter’,” said Delor, while noting it is a rare balance to strike. Promising candidates with all three criteria include 2D halide perovskites, minerals increasingly used in solar panels and LEDs, and a class of 2D semiconductors known as transition-metal dichalcogenides, or TMDs.

Earlier this year, Delor and his colleagues used a TMD to show how polaritons can enhance nonlinear optical interactions in waveguides. These are structures that confine and direct light within a material, and are highly compatible with silicon-based chips, such as those used in emerging optical computers. With what they’ve learned about producing polaritons in the current work, they now plan to optimize polariton-enhanced nonlinear interactions in waveguides with the goal of using light to change the properties of single photons. This would create a quantum version of a computer gate, from light—a key step towards realizing light-based quantum computing architectures.

“Enhancing optical interactions to be sufficient for single-photon nonlinearities is a tremendous challenge, but one that could immediately unlock countless applications in quantum information and sensing,” said Delor. “We think that these optimized polaritons are a highly promising and scalable approach to reaching this grand scientific goal.”

END