🔬 optics

Field-Widened Multimode Interferometer with Long Time-Bin Delay Using a Multi-Pass Herriott Cell

This paper presents and validates a passive, field-widened multimode interferometer utilizing a multi-pass Herriott cell to achieve a compact design with a 12ns time-bin delay and a $0.4^{\circ}$ field-of-view, enabling high-visibility interference for spatially multimode optical signals in free-space channels without adaptive optics.

Original authors: Ramy Tannous, Stéphane Vinet, Kaylee Sherk, Kimia Mohammadi, Thomas Jennewein

Published 2026-08-14

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ramy Tannous, Stéphane Vinet, Kaylee Sherk, Kimia Mohammadi, Thomas Jennewein

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to send a secret message using flashes of light. In the world of quantum physics, these flashes can carry information in a special way called "time-bin encoding," where the message is hidden in when the flash arrives, not just how bright it is. This is a favorite trick for sending quantum secrets through fiber optic cables, like the ones in your internet. But what if you want to send these secrets through the open air, like from a satellite to a ground station? That's where things get messy. The air is full of turbulence, like heat waves rising off a summer road, which distorts the light beam. It's like trying to aim a laser pointer at a moving target while standing on a wobbly boat.

To fix this, scientists usually use fancy, active systems that act like a "smart mirror" to correct the wobble in real-time. But these systems are heavy, complex, and can lose a lot of the precious light signal. A simpler, "passive" solution is to build a special kind of light splitter called an interferometer that is so forgiving it doesn't care if the light beam is wiggling or distorted. The catch? To make these "field-widened" interferometers work for long delays (waiting for the light to travel a long way), they usually need to be huge, taking up the space of a whole room, or they need to be made of heavy, dense glass that is hard to work with.

This is the puzzle a team of researchers from Canada set out to solve. They wanted to build a light-splitting machine that could handle long delays, work with messy, distorted light beams, and still fit into a small box. They didn't just dream it up; they built a prototype and proved it works.

The team, led by Ramy Tannous and colleagues, designed a clever device that uses a "Herriott cell." Think of a Herriott cell as a light trap made of two curved mirrors facing each other. Instead of letting light bounce straight back and forth, they arrange the mirrors so the light bounces in a spiral pattern, hitting the mirrors many times before escaping. It's like a pinball machine where the ball bounces around the table dozens of times, traveling a very long distance, all while staying inside a small, compact room.

By folding the light path into this spiral, the researchers could create a massive delay—making the light wait for 12 nanoseconds (that's 12 billionths of a second)—without needing a device that is meters long. Usually, a 12-nanosecond delay would require a path length difference of about 3.6 meters (since light travels roughly 30 centimeters per nanosecond). By using this multi-pass trick, they managed to squeeze that long journey into a much smaller footprint.

The real magic, however, is that this compact design is also "field-widened." In plain English, this means the device is incredibly forgiving. If the light beam enters at a slightly weird angle or is distorted by the atmosphere, the device still works perfectly. The researchers tested this by shooting light through their device in two ways: first with a perfect, smooth laser beam, and second with a messy, "multimode" beam that had been scrambled to look like light traveling through a turbulent atmosphere.

The results were impressive. In their simulations and with their physical prototype, the device maintained a "visibility" (a measure of how well the light waves interfere to create a clear signal) of over 0.9, even with the messy beams. When they tested it with a real laser, they measured a visibility of 0.97 for smooth light and 0.95 for the messy light. Even when they used a pulsed laser to create actual time-bin messages and sent them through a long, distorted fiber optic cable before hitting their device, they still achieved a visibility of 0.88.

The paper explicitly rules out the idea that you need massive, room-sized setups or heavy, dense materials to get these long delays with wide acceptance angles. They showed that the standard, uncorrected interferometers (the kind without the special mirror tricks) fail miserably when the light angle changes even a tiny bit, causing the signal to drop. Their Herriott cell design, however, kept the signal strong even as they tilted the incoming light.

They are very sure about these findings because they didn't just rely on computer models; they built a physical prototype using standard mirrors and tested it with real lasers at different colors (532 nm and 785 nm). While they noted that the visibility dropped slightly in the pulsed laser test compared to the continuous laser test—likely due to small imbalances in their equipment or the long path losses—they confirmed that the device successfully handles the 12-nanosecond delay and the messy light conditions.

In short, the team demonstrated a way to build a "smart" light splitter that is small enough to fit on a satellite, robust enough to handle the wobbly atmosphere, and fast enough to keep quantum secrets safe. It's a step toward making quantum communication through the air much more practical, proving that you don't need a giant machine to do big science.

Technical Summary: Field-Widened Multimode Interferometer with Long Time-Bin Delay Using a Multi-Pass Herriott Cell

Problem Statement
Interference of optical signals in free-space channels is frequently compromised by atmospheric turbulence, which distorts spatial modes and necessitates complex, active adaptive optics systems to maintain coupling into single-mode fibers. While field-widened interferometers offer a passive alternative by increasing the accepted solid angle, implementing them with long time-bin delays (required for interfacing with stationary qubits like trapped ions or solid-state devices) presents significant challenges. Conventional free-space interferometers with large path length differences become physically bulky and difficult to stabilize. Furthermore, standard interferometers suffer from decreased interference visibility when subjected to spatial mode distortions or dynamic fluctuations in the optical path length difference caused by pointing errors and turbulence.

Methodology
The authors propose a compact, passive field-widened interferometer design utilizing a multi-pass Herriott cell (HC) to achieve long optical path delays within a small form factor. The design is conceptualized in a Michelson configuration but is adaptable to Mach–Zehnder layouts.

Design and Optimization: The team employed a 6×6 ray transfer matrix method to model the non-sequential 3D ray tracing of the system. This allowed for the optimization of the Herriott cell parameters—specifically the mirror radii of curvature ( $R_1, R_2$ ) and the mirror separation ( $\ell$ )—to ensure that the output rays of the long-delay path overlap spatially and directionally with the short-delay path, regardless of the input angle of incidence. This optimization targets a condition where the optical path difference has a minimized second-order derivative with respect to the input angle.
Verification: To account for wavefront distortions, polarization, and physical constraints (such as aperture clipping) not captured by the matrix method, the authors utilized a non-sequential ray tracing package based on Gaussian Beamlet Decomposition (GBD). This simulation verified the interference visibility for both single-mode Gaussian inputs and highly structured multi-mode wavefronts across various angles of incidence.
Experimental Prototype: A proof-of-principle prototype was constructed using standard bulk optics and optomechanics. The device utilized 2-inch spherical mirrors with a custom entrance aperture (a 3 mm hole). The system was tested using continuous-wave (CW) lasers at 532 nm and 785 nm, as well as attenuated pulsed lasers at 785 nm. The experimental setup included a 5-meter multi-mode fiber to intentionally distort the spatial mode of the input beam before it entered the HC interferometer.

Key Contributions

Compact Long-Delay Architecture: The paper demonstrates a method to fold long optical path lengths (corresponding to nanosecond-scale time delays) into a compact footprint using a multi-pass Herriott cell, addressing the bulkiness of traditional free-space delay lines.
Multimode Compatibility: The design is explicitly engineered to maintain high interference visibility for spatially multimode beams, a critical requirement for free-space quantum communication where atmospheric turbulence degrades spatial coherence.
Field-Widened Performance: The optimized geometry ensures that the interferometer remains robust against variations in the angle of incidence, a common issue in mobile free-space links.

Results

Simulation: GBD simulations confirmed that the optimized Herriott cell designs (e.g., with $R_1=1000$ mm, $R_2=700$ mm, and 18 bounces) achieve interference visibilities greater than 0.99 for both single-mode and multi-mode inputs. The visibility remained high across a range of input angles until beam clipping occurred at the entrance aperture.
Continuous Wave Testing: Experimental tests with CW lasers at 532 nm and 785 nm yielded high visibilities. For the multi-mode input, the measured visibility was 0.970 at 532 nm and 0.951 at 785 nm.
Pulsed Time-Bin Testing: Using an attenuated pulsed laser and a multi-mode fiber to simulate a distorted channel, the prototype successfully interfered time bins with a path length difference of 12 ns. The measured superposition basis interference visibility was 0.88. The authors attribute the slight reduction in visibility compared to CW results to beam splitter imbalances, higher losses in the long paths, and potential mismatches in time delay between the fiber-based delay and the HC interferometer.

Significance and Claims
The authors claim that this work presents the longest interferometer path difference achieved for a field-widened interferometer to date. The design is significant because it provides a passive, robust solution for quantum communication systems that require long time-bin delays and must operate over multimode free-space channels without the overhead of active adaptive optics. The compact form factor makes the architecture particularly attractive for resource-constrained platforms, such as nano-satellites. The authors note that while their current analysis is limited to single-plane Herriott patterns, the design principles could be extended to three-dimensional patterns for even longer delays or integrated into frequency-bin and hybrid time-frequency quantum communication protocols. The demonstrated visibility is deemed sufficient for applications such as quantum key distribution and entanglement swapping.

More like this