🔭 astrophysics

Halo stars harbour few wide ultracool companions

This study presents the first search for wide ultracool companions to metal-poor halo stars, finding no such companions and establishing a strict upper limit of 4.0% on their occurrence rate, which suggests they are intrinsically rare and likely form or persist less efficiently than wide stellar companions in low-metallicity environments.

Original authors: Jerry Jun-Yan Zhang, Nicolas Lodieu, Eduardo L. Martín

Published 2026-06-02

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Jerry Jun-Yan Zhang, Nicolas Lodieu, Eduardo L. Martín

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 the Milky Way galaxy as a massive, bustling city. Most of the stars we see are like the modern, shiny skyscrapers in the downtown district—they are rich in heavy elements (like iron and carbon) and relatively young. But scattered in the outskirts are the "old neighborhoods," known as the halo. These areas are home to ancient, "metal-poor" stars that formed when the universe was just a toddler. They are like the weathered, brick cottages of the galaxy, made mostly of the pristine, light materials from the Big Bang.

Astronomers have long wondered: Do these ancient, metal-poor stars have "pets"?

Specifically, do they have wide companions that are Ultracool Dwarfs? Think of these dwarfs as the "tiny, dim fireflies" of the stellar world. They are so cool and faint that they barely glow, sitting on the border between being a star and a planet. Because they are so old, they have had billions of years to cool down, making them incredibly hard to spot.

The Great Cosmic Search

The team behind this paper, led by J.-Y. Zhang, decided to play a game of "Cosmic Hide-and-Seek." They wanted to see if these ancient metal-poor stars had any of these dim, wide-orbiting fireflies as companions.

The Strategy:

The Target List: They picked 66 of the nearest, brightest, and fastest-moving ancient stars. These stars are like lighthouses; because they move fast across the sky, it's easier to tell if something else is moving with them (a companion) or just passing by in the background.
The Camera: They used a giant telescope in the Canary Islands (the GTC) equipped with a super-sensitive camera that sees in infrared light (heat). This is like using night-vision goggles to spot the faint glow of a firefly in the dark.
The Two-Step Dance: They took a picture of these stars in 2020/2021, and then came back in 2024 to take another picture. By comparing the two, they could see if any faint dots moved along with the main star. If a dot moved with the star, it's a companion. If it stayed still or moved differently, it was just a background star.

The Results: A Quiet Neighborhood

After all that effort, the result was surprisingly quiet.

The Fireflies: They found zero confirmed wide ultracool companions. Not a single one.
The Pets: They did find four companions that were actually stars (not the tiny dwarfs they were looking for). This means about 6% of the ancient stars they checked have a stellar "buddy."
The Limit: Because they didn't find any of the tiny dwarfs, they calculated an "upper limit." It's like saying, "If there are any of these tiny companions hiding, they are so rare that fewer than 4 out of every 100 ancient stars have one."

Why So Empty?

The paper suggests a few reasons why these ancient stars might be lonely:

The "Cooling" Problem: These companions are so old (billions of years) that they have cooled down so much that they have turned into spectral types that are too faint for our current telescopes to see. It's like trying to find a candle that has burned down to a tiny, invisible ember. The only ones we could potentially see are the ones that are in a very narrow, transitional phase, which are naturally rare.
Formation Issues: It might be that nature just doesn't like making these specific pairings. Forming a tiny, dim companion next to a big, old star might be a difficult process, like trying to balance a feather on a boulder.
The Cosmic Shuffle: Over billions of years, the galaxy is a busy place. Other stars passing by might have kicked these wide companions out of their orbits, leaving the ancient stars alone.

The Bottom Line

The paper concludes that while ancient, metal-poor stars do have some stellar companions, they almost never have these specific wide, ultra-cool "firefly" companions.

The authors also note something interesting: It doesn't seem to matter how "metal-poor" the star is. Whether the star is very old or slightly less old, the frequency of these companions stays low. This suggests that the rules for forming these systems are strict, regardless of the star's chemical makeup.

In short: The ancient outskirts of our galaxy are surprisingly quiet when it comes to these specific types of dim, wide-orbiting companions. They are rare, elusive, and perhaps a bit lonely.

Technical Summary: Halo stars harbour few wide ultracool companions

Problem and Motivation
Ultracool dwarfs (UCDs) with spectral types later than M7.0 and effective temperatures below 2700 K serve as valuable fossil tracers of the early Milky Way, particularly when found in metal-poor halo populations. While several specific systems of wide halo ultracool companions have been identified (e.g., HD 114762B, Wolf 1130C), the statistical frequency of such wide ultracool companions around metal-poor halo stars remains poorly constrained. Previous studies have investigated wide stellar companions to metal-poor stars, finding frequencies ranging from a few percent to ~20%, and have studied wide ultracool companions around solar-metallicity and young stars. However, the specific frequency of wide ultracool companions around metal-poor halo stars, and how metallicity influences the formation and retention of these systems, has not been previously quantified. This study aims to fill that gap by providing the first wide ultracool companion frequency for metal-poor halo stars.

Methodology
The authors conducted a dedicated survey targeting 66 nearby, bright halo stars with spectroscopically determined metallicities of [Fe/H] < −1.5 dex. The sample was drawn from the Carney et al. (1994) catalogue, ensuring high proper motions and reliable membership. All targets were within 250 pc, with effective temperatures between 4600 and 6300 K.

The observation strategy involved two epochs of deep near-infrared (NIR) imaging:

First Epoch (2020–2021): Deep J-band imaging was performed using the EMIR instrument on the 10.4-m Gran Telescopio Canarias (GTC). The survey utilized a seven-point dithering pattern with a total on-source exposure of 1260 s per object.
Candidate Selection: First-epoch images were visually compared with Pan-STARRS optical images (y, z, i bands). Sources detected in the deep J-band but not in Pan-STARRS (or showing significant motion relative to the primary) were flagged as candidates.
Second Epoch (2023–2024): 21 targets exhibiting faint, non-Pan-STARRS sources in the first epoch were re-observed with GTC/EMIR+ (an upgraded detector) to establish a baseline of two to four years.
Analysis: The team searched for comoving sources between the two epochs using relative astrometry. They performed aperture photometry to determine limiting magnitudes and calculated the completeness of the survey for different spectral types of extreme subdwarfs (esd) based on known benchmarks like WISE1810−1010 and Wolf 1130C.

Key Results

Detection Limits: The survey reached average limiting magnitudes of $J_{lim} = 22.8$ mag (first epoch) and $J_{lim} = 23.0$ mag (second epoch). This depth enabled the detection of extreme subdwarfs earlier than type esdT0 at distances up to 250 pc.
Ultracool Companions: No bona fide wide ultracool comoving companions were found among the 57 independent trials (66 targets minus 2 with insufficient first-epoch depth and 7 without second-epoch follow-up). One candidate (around BD+02°3375) was identified but rejected as a comoving companion because its proper motion differed significantly from the primary (6 $\sigma$ difference).
Stellar Companions: Four wide stellar companions were confirmed via Gaia data within the sample.
Frequency Limits:
- The upper limit for the frequency of wide ultracool companions (spectral types earlier than esdT0) to metal-poor halo stars is 4.0% at a 90% confidence level.
- The derived frequency for wide stellar companions in the same sample is 6.1 $^{+7.2}_{-4.0}$ % at a 90% confidence level.
Completeness: The survey was complete for extreme subdwarfs with spectral types earlier than esdT0 within 250 pc. Sensitivity was insufficient to assess the presence of late-type esdTs or esdYs across the majority of the sample.

Significance and Conclusions
The study concludes that wide ultracool companions are rare around metal-poor halo stars, with a frequency marginally lower than that of wide stellar companions. The authors assert that, given current uncertainties, there is no statistically significant evidence for a dependence of wide ultracool companion frequency on metallicity, separation, or primary mass.

The paper suggests that the scarcity of these systems may be attributed to two factors:

Formation Efficiency: Formation and retention processes in binary systems may operate less efficiently for ultracool secondaries, particularly those with extreme mass ratios.
Evolutionary Cooling: For the halo population, most brown dwarfs may have cooled over ~10 Gyr to very late spectral types (beyond esdT0), falling below the detection limits of the survey. The observed upper limit likely reflects only the small fraction of transitional brown dwarfs occupying a narrow mass range.

The authors note that the very low upper limit implies that survey-driven strategies using deep NIR surveys (such as Euclid and Roman) are more effective than continued deep imaging of individual targets for constraining these frequencies.

The Great Cosmic Search

The Results: A Quiet Neighborhood

Why So Empty?

The Bottom Line

More like this