Monday, 13 July 2026CurrentDeck — Live news signals. Clear context.
CDCurrentDeck
Signal-rich news, source-grounded context
Tech & Science

MIT researchers develop tiny infrared chip for tunable gas and heat sensing

This new metasurface-based device uses a crossbar architecture to independently control infrared light, offering a scalable solution for environmental monitoring and medical diagnostics.

MIT researchers develop tiny infrared chip for tunable gas and heat sensing
MIT researchers develop tiny infrared chip for tunable gas and heat sensing

Mit researchers have developed a novel chip-based optical device capable of dynamically controlling infrared light, opening new possibilities for gas detection, thermal imaging, and environmental monitoring. The technology, described in a study published in *Nature Communications*, uses a tunable lens made of microscopic pixels that independently manipulate incoming infrared light, eliminating the need for mechanical components and enabling compact, scalable systems.

The device leverages metasurfaces — thin, engineered materials with nano-scale patterns that interact with light. Unlike earlier metasurfaces, which altered optical properties across entire surfaces, the MIT team’s design allows individual pixel control. This innovation was achieved using a crossbar architecture inspired by display technologies, where two layers of copper wires intersect beneath a doped silicon layer. At each intersection, heat generated by the silicon switches a phase-change material between crystalline and amorphous states, modifying how it interacts with mid-infrared light. A built-in diode selector prevents electrical interference between pixels, enabling reliable operation over thousands of cycles.

The researchers fabricated a 6-by-6 metasurface pixel array using MIT.nano facilities and a semiconductor chip foundry, demonstrating the system’s scalability. “This architecture allows us to scale to potentially millions of pixels without current leakage issues,” said Juejun Hu, MIT’s John F. Elliott Professor of Materials Science and Engineering. The team aims to expand the pixel count and refine the design for industrial applications, emphasizing compatibility with existing chip manufacturing processes.

The mid-infrared wavelength range — invisible to the human eye but critical for detecting heat signatures and molecular absorption, is central to the device’s functionality. Many organic molecules, including methane and propane, absorb mid-infrared light, making the chip valuable for monitoring air quality, identifying gas leaks, and studying atmospheric compounds. “This could give us more information as we study space or help with environmental protections,” said first author Cosmin-Constantin Popescu PhD ’25.

Potential applications extend beyond environmental sensing. The technology could enhance thermal imaging for military night vision, improve medical diagnostics by detecting tissue abnormalities, and enable optical computing. By encoding neural network weights into metasurfaces, the device could process light-based computations, though practical implementations remain in early stages. “Researchers have already used this approach to emulate complex neural networks,” Hu noted.

The work builds on MIT’s prior research into phase-change metasurfaces, including a 2021 lens that adjusted focus through heat-induced material shifts. The new design overcomes limitations of earlier systems by achieving two-dimensional pixel-level control, a milestone described as “the first time anyone’s implemented it” in the field. Collaborators included researchers from the University of Central Florida, University of Washington, and Korea Advanced Institute of Science and Technology, alongside industry partners like Dynasil and Draper Laboratory.

While the chip’s immediate impact lies in compact, high-precision sensing, its long-term potential hinges on scaling and integration. The team is exploring ways to tailor the system for specific applications, such as highlighting known molecular signatures in real-time. “You might be looking for a human in a dark room or specific features in an image, and that prior information can be useful,” Hu said. The research was supported by the U.S. Air Force, National Science Foundation, and other institutions, underscoring its strategic importance for defense, environmental, and scientific sectors.

Related stories