A fingernail-sized chip just let two telescopes see as one

A glass chip small enough to rest on a fingertip has, for the first time, successfully combined starlight collected by two separate telescopes well enough to let them function as a single, far more powerful instrument, researchers at Heriot-Watt University in Edinburgh reported. The device was tested during on-sky observing campaigns in 2024 and 2025 at the CHARA Array, a six-telescope observatory on Mount Wilson in California operated by Georgia State University, and the results have now been published in the journal Monthly Notices of the Royal Astronomical Society.

The technique the chip performs, called interferometry, is standard practice in astronomy: by combining light collected from telescopes spread across some distance, astronomers can resolve detail equivalent to a single telescope as wide as the separation between them. The CHARA Array's six 1-meter telescopes, arranged across baselines of up to 331 meters, already work this way. What is new is the hardware doing the combining. Where interferometers have traditionally relied on elaborate arrangements of mirrors, lenses and precisely aligned optical paths to merge the beams, the Heriot-Watt chip performs the same function inside a block of specialized glass only a few centimeters long.

What the numbers show

The chip was built using a technique called ultrafast-laser inscription, in which pulses from a laser etch light-guiding channels directly into a type of low-OH Infrasil glass, the same basic approach used to make integrated circuits guide electrons rather than photons. It sits at the core of an instrument called CHARIOT — short for the CHARA Array Integrated Optics Testbench — which was designed so that different experimental chips can be swapped in and out within hours, letting the team iterate on designs quickly.

In its first on-sky run, the paper reports, CHARIOT measured the raw squared visibilities used to characterize stars with sub-1 percent precision for targets around magnitude 2 in the K band or brighter — a level of accuracy the authors say is comparable to JouFLU, the conventional instrument the chip is designed to eventually replace. Under poor-to-moderate atmospheric seeing conditions, the system also detected interference fringes from a star as faint as magnitude 5.212, fainter than the older setup typically reaches. The research collaboration, which also includes the Leibniz Institute for Astrophysics Potsdam and the University of Cologne in Germany, described the work as a milestone in a research program that has been underway for more than 15 years.

Why a smaller box of glass matters

The appeal of replacing bulky optics with a chip is partly economic and partly practical. Interferometric beam combiners built from discrete mirrors and lenses require painstaking alignment and are sensitive to vibration and temperature changes, problems that tend to multiply as more telescopes and light paths are added to a system. A chip-based combiner, by contrast, is intrinsically stable once fabricated, takes up a fraction of the space, and, proponents argue, could ultimately be manufactured at lower cost, an important consideration given that observing time on major facilities is itself expensive.

Robert Thomson, a professor at Heriot-Watt's Institute of Photonics and Quantum Sciences who co-developed the chip with Aurélien Benoît, said the result validates an idea the group has pursued for years. "This is a significant milestone for a research programme that began more than 15 years ago," Thomson said, adding that the work shows "much of that work can instead be carried out by a glass chip only a couple of centimetres in size."

"The great advantage of our approach is that we can integrate complex optical functions into a remarkably small and stable glass chip," said Aurélien Benoît, a researcher on the project, describing the device as a flexible platform for testing advanced photonic technologies.

Context: a broader push toward photonics in astronomy

The Heriot-Watt result is part of a wider trend of astronomers turning to integrated photonics to replace conventional optics in sensitive instruments. A similar chip-based approach already underpins GRAVITY, an instrument at the European Southern Observatory's Very Large Telescope Interferometer in Chile that has been used to study the supermassive black hole at the center of the Milky Way. A follow-on instrument called PLANETES, under development at the Paris Observatory, aims to extend that approach to shorter infrared wavelengths and is expected to be installed on the VLTI's four telescopes by the end of 2028. Separately, Heriot-Watt's underlying chip-fabrication method has already been commercialized for fiber-optic data communications through Optoscribe, a spin-out co-founded by Thomson, a sign that the underlying manufacturing process is mature even as its astronomical application is new, according to a summary of the work by the trade outlet Electro Optics.

Significance and next steps

The CHARIOT demonstration used only two of the CHARA Array's telescopes and tested a single type of chip design; the authors frame the published paper as the first in a planned series documenting the instrument's performance. The research team says CHARIOT's swappable design means it will continue to serve as a testbed on which improved chip designs, including ones capable of combining light from more than two telescopes at once, can be validated on real starlight before being adopted elsewhere.

Aline Dinkelaker of the Leibniz Institute for Astrophysics Potsdam, the paper's lead author, said the long-term payoff could extend well beyond making existing interferometers more compact. "As the technology develops, it could help astronomers study planetary systems around other stars in greater detail," Dinkelaker said, noting that the on-sky results demonstrate that "photonic technologies can deliver the performance needed for frontline astronomical research."

Directly imaging planets around other stars — distinguishing a faint planet from the overwhelming glare of its host star — is among the most technically demanding goals in observational astronomy, and interferometry is one of the few techniques capable of delivering the angular resolution required. Whether chip-based combiners like CHARIOT's eventually find their way into purpose-built exoplanet-imaging instruments will depend on further on-sky tests, including attempts to combine light from more than two telescopes simultaneously and to extend performance to fainter targets. The research was supported by the UK's Science and Technology Facilities Council and the European Union, according to the researchers.