Optical system taps nanostructured birefringent metasurface to detect atmospheric turbulence

As satellites, military networks, and next-gen telecom applications shift from fiber infrastructure to high-speed optical links through the air, it exposes the signal to atmospheric turbulence.

To detect these distortions in real time, CFD Research created a low-latency optical system that features a nanostructured birefringent metasurface. It’s a big deal because free-space optical communications systems must be as latency-free and reliable as possible.

CFD Research’s Arturo Martin Jimenez, a research engineer for the Advanced Optics Branch, and Zachary Coppens, manager and senior research engineer of the Advanced Optics Branch, answer Laser Focus World’s questions about their low-latency optical system.

Laser Focus World: What inspired your work with metasurfaces and free-space optics?

CFD Research: CFD Research’s R&D team is driven by both curiosity and a desire to make a real impact. We spend a lot of time finding problems that matter most and digging past the surface to understand the root challenges. Free-space optics (FSO) caught our attention because of its growing importance and the fact that metasurface technologies, which is one of our core competencies, could potentially help. Throughout our discussions with the community, we found that conventional approaches to wavefront sensing fall short in deep turbulence, which can limit FSO range. So we came up with a solution using metasurfaces to overcome this limitation, and it’s rewarding to feel that our contributions might help the development and implementation of FSO systems.

LFW: Can you describe the metasurface optics involved?

CFD Research: Our approach uses metasurface optics to reduce the size and latency of a conventional phase-diversity wavefront sensor. The metasurface features birefringent meta-atoms in a super-cell architecture that creates multiple channels through which we can encode the signal. Birefringence enables polarization multiplexing, which imparts a different phase to orthogonal polarizations of light. And the super-cell design allows for spatial multiplexing at the focal plane array (FPA) and diffracts light to four separate orders, each with a different phase profile. In total, we can achieve eight signal channels with our metasurface design.

LFW: Benefits of a nanostructured birefringent metasurface?

CFD Research: Conventional phase-diversity wavefront sensors are robust against deep turbulence conditions. But their implementations require bulky systems or moving parts that introduce latency. Using a birefringent metasurface enables the capture of multiple intensity measurements corresponding to various phase diversities on a single detector. This significantly reduces the size of the sensor and eliminates the need for any mechanical movement or sequential phase modulation typically used by conventional systems to make the measurements.

The birefringent properties of the metasurface are engineered by breaking the symmetry of the nanostructures that compose the device. These nanostructures, smaller than the wavelength of light, function as meta-atoms whose electromagnetic response can be tailored to exhibit properties not inherent to the bulk material. Such design flexibility is a unique advantage metasurfaces provide when compared to conventional optical technologies.

LFW: Most surprising/coolest aspects of your work?

CFD Research: There were a few ‘aha’ moments throughout this project. The first came when we were designing the metasurface and reconstruction algorithm through simulations. We were first optimizing the algorithm to give us a reconstruction phase matching the phase coming into the system. This was working well for the most part, but the trick about deep turbulence is that it introduces weird things to the phase profile of the beam, such as branch points, which make it difficult to use quantitative metrics of error like root mean squared error (RMSE) to determine the reconstruction accuracy.

Fortunately, we realized that for FSO, what we really care about is how much we can maximize the signal at the target—and we adjusted our approach to instead optimize this signal after correcting with the predicted phase. It took more effort than we expected to implement this, but it was worth it because the performance increased noticeably. We were very excited when the results came back. Another moment came later in the lab. There’s always some nervous energy before testing, and you wonder if all those careful simulations will hold up in the real world. When we finally compared the experimental data with our simulated predictions and saw how closely they matched, it was incredibly satisfying. These are the moments that remind us why we love this kind of work.

LFW: Any technical challenges to overcome?

CFD Research: As we work to raise the technology readiness level and move our system into outdoor testing, we know new technical challenges will come with it. That’s part of the fun. Increasing system fidelity always brings hurdles—in our case, we anticipate challenges with signal-to-noise ratio, ruggedization, system calibration, and achieving real-time performance on embedded hardware. Each of these areas pushes us to grow our understanding and refine the design. But we’re genuinely excited to take on these challenges because every step forward brings us closer to seeing the technology operate in the real world.

LFW: What does your work mean for telecom, satellites, and military applications?

CFD Research: This is an area we can’t go into detail on, but broadly speaking the technology supports improved reliability and performance across a range of optical systems.

LFW: Timeline to use/what’s next?

CFD Research: This technology has already sparked a lot of enthusiasm within the community, which is really encouraging. Right now, we’re in discussions with several partners about maturing it through a phased approach where we start with outdoor testing within a ruggedized enclosure and ultimately move toward platform integration. CFD Research is eager to keep pushing this forward because real impact only happens once the technology makes its way into the field.

FURTHER READING

A. Martin Jimenez, M. Baltes, J. Cornelius, N. Aközbek, and Z. J. Coppens, Nat. Photon. (2025); https://doi.org/10.1038/s41566-025-01772-4.