High CO/H2 ratios supports an exocometary origin for a CO-rich debris disk

By using CRIRES+ near-infrared spectroscopy to detect strong CO absorption but no H$_2$ in the edge-on debris disks of HD 110058 and HD 131488, the study establishes high CO/H$_2$ ratios that rule out a primordial origin and strongly support an exocometary source for the gas.

The Gist

Imagine you are a cosmic detective trying to solve a mystery about the "atmosphere" of a young star system.

The Mystery: Where did the gas come from?

In our solar system, we have a dusty ring of rocks and ice left over from when the planets were forming. Astronomers call these debris disks. For a long time, scientists thought these disks were just dusty and empty of gas. But recently, they found a lot of Carbon Monoxide (CO) gas in many of these disks.

This created a big puzzle with two possible suspects:

1. The "Primordial" Suspect: This gas is the "leftover" from the giant cloud of gas and dust that formed the star billions of years ago. It's like the original soup the star was cooked in. If this is the case, the gas should be mostly Hydrogen (H2) with a little bit of CO mixed in (like a huge bowl of water with a few drops of soda).
2. The "Exocometary" Suspect: This gas is fresh. It's being released right now by icy comets crashing into each other or melting as they get close to the star. Think of it like a giant, icy snowball melting in the sun. If this is the case, the gas should be mostly CO and very little Hydrogen (like a bowl of pure soda with almost no water).

For years, astronomers could see the CO (the soda), but they couldn't find the Hydrogen (the water). Without knowing how much water was there, they couldn't tell which suspect was guilty.

The Investigation: Looking for the "Ghost"

The authors of this paper decided to play detective using a very powerful telescope called the VLT (Very Large Telescope) equipped with a super-sharp camera called CRIRES+.

They looked at two specific star systems, HD 110058 and HD 131488. These stars are about 15 million years old (very young in cosmic terms) and have disks that are perfectly edge-on, like looking at a plate from the side. This is perfect for spotting gas because the star acts like a bright flashlight shining through the gas, making the gas cast a shadow (an absorption line) that the telescope can see.

They were hunting for two things:

• CO: They knew this was there.
• H2 (Hydrogen): This is the "ghost" they were trying to catch. If the gas was primordial (leftover soup), there should be a lot of Hydrogen. If it was exocometary (melting comets), there should be almost none.

The Clue: The Missing Hydrogen

The telescope was so sensitive it could have seen even a tiny whisper of Hydrogen. But here is the twist: They found the CO, but they found absolutely no Hydrogen.

It's like walking into a room where you can smell strong coffee (CO), but you can't find a single drop of water (H2) anywhere.

The Verdict: The "Exocometary" Theory Wins

Because they couldn't find the Hydrogen, the scientists calculated the ratio of CO to H2.

• Primordial Gas (Leftover Soup): Should have a ratio of about 1 CO for every 10,000 Hydrogens.
• What they found: In one of the stars (HD 110058), there was so much CO and so little Hydrogen that the ratio was 1,000 times higher than what you'd expect from leftover soup.

The Analogy:
Imagine you find a puddle on the sidewalk.

• Scenario A (Primordial): You think the puddle is from a leaky fire hydrant that has been dripping for 100 years. You expect it to be mostly water with a little bit of rust.
• Scenario B (Exocometary): You think the puddle is from someone dumping a bucket of soda. You expect it to be mostly soda with almost no water.

When you taste the puddle, it's 99% soda. You can conclude with high confidence that it wasn't a slow leak from a hydrant; someone dumped a bucket of soda.

Why This Matters

1. It's Fresh Gas: The gas in these disks isn't ancient leftovers. It is being constantly replenished by icy comets (exocomets) crashing and melting.
2. The "Shielding" Problem: Usually, if there's a lot of gas, the Hydrogen acts like a shield, protecting the CO from being destroyed by the star's UV light. Since there is no Hydrogen, the CO should be destroyed quickly. The fact that it's still there means the comets must be releasing gas very fast to keep the supply full.
3. Planetary Birth: These systems are at the age where Earth-like planets are forming. If comets are melting and releasing gas and ice right now, they might be delivering the water and ingredients needed for life to these new planets.

The Bottom Line

This paper solved a decades-old mystery for at least one of these star systems. By proving there is almost no Hydrogen, they confirmed that the gas we see isn't the "old soup" from the star's birth. Instead, it is a fresh, active "soda fountain" being sprayed by a swarm of icy comets, giving us a new way to understand how planetary systems evolve and how water might get to new worlds.

Technical Summary

Here is a detailed technical summary of the paper "High CO/H2 ratio supports an exocometary origin for a CO-rich debris disk" by Smith et al. (2026).

1. Problem Statement

Over 20 exocometary belts (debris disks) have been found to host detectable circumstellar gas, primarily in the form of carbon monoxide (CO). A fundamental ambiguity exists regarding the origin of this gas:

• Primordial Origin: The gas is a remnant of the parent protoplanetary disk. In this scenario, the gas should be rich in molecular hydrogen ($H_2$), with a CO/$H_2$ ratio typically between $10^{-4}$and$10^{-6}$, similar to the interstellar medium (ISM) and young protoplanetary disks.
• Secondary (Exocometary) Origin: The gas is produced in-situ via outgassing from colliding comets. In this scenario, the gas is expected to be $H_2$-poor, as $H_2$ is not a primary volatile in cometary ices (unlike CO), leading to significantly higher CO/$H_2$ ratios.

While the origin of gas in CO-poor belts has been largely resolved as secondary due to insufficient $H_2$ shielding, the origin of gas in CO-rich belts remains an open question. Theoretical models suggest that if these belts were primordial, the abundant $H_2$ would shield CO from UV photodissociation, allowing it to survive for the system's age (~15 Myr). However, if the gas is secondary, the lack of $H_2$ shielding implies the CO must be replenished rapidly. Direct measurement of $H_2$ column densities in CO-rich belts has been lacking, preventing a definitive distinction between these scenarios.

2. Methodology

The authors aimed to break this dichotomy by directly measuring $H_2$ column densities in two edge-on, CO-rich debris disks around $\sim$15 Myr-old A-type stars: HD 110058 and HD 131488.

• Observations:

  • Instrument: High-resolution near-infrared spectrograph CRIRES+ on the Very Large Telescope (VLT).
  • Targets: HD 110058 and HD 131488 (inclinations of $\sim$85.5° and 82°, respectively), chosen for their high CO column densities and edge-on geometry, which maximizes the line-of-sight column density against the stellar continuum.
  • Spectral Range: K-band (1904–2452 nm).
  • Key Transitions:
    • $H_2$: The rovibrational absorption line $v=1-0$ S(0) at 2223.3 nm (ground state).
    • $^{12}CO$: The $v=2-0$ rovibrational band (2333.8–2335.5 nm).

• Data Reduction & Analysis:

  • Reduction: Used the ESO CRIRES+ pipeline (ESOREX) with custom flat-fielding and nodding subtraction.
  • Telluric Correction: Applied the molecfit software (v4.3.1) to remove atmospheric absorption, which is critical for isolating the weak $H_2$ signal.
  • Calibration: Addressed "super-resolution" effects caused by the slit not being fully illuminated, requiring separate wavelength calibration for nodding positions to ensure line alignment.
  • Modeling:
    • Used RADIS (a line-by-line spectral modeling tool) to generate synthetic absorption spectra.
    • $H_2$: Assumed Local Thermodynamic Equilibrium (LTE) and a single column density/temperature. Since $H_2$ was not detected, the model provided upper limits on column density.
    • $CO$: Modeled using a simplified non-LTE approach (distinct kinetic and excitation temperatures) to fit the detected absorption lines.
    • Statistical Analysis: Employed Markov-Chain Monte Carlo (MCMC) methods (emcee) to derive posterior probability distributions for column densities and temperatures, calculating 3$\sigma$ lower limits for the CO/$H_2$ ratio.

3. Key Contributions

• First Direct Probe: This study presents the first direct spectroscopic probe of $H_2$ gas in CO-rich exocometary belts using near-IR absorption spectroscopy.
• Methodological Advancement: Demonstrates the efficacy of CRIRES+ in detecting faint rovibrational absorption lines in debris disks, overcoming telluric contamination and instrumental systematics through rigorous nodding and molecfit correction.
• Compositional Constraints: Provides the first empirical constraints on the CO/$H_2$ abundance ratio in CO-rich debris disks, moving beyond indirect kinematic or shielding arguments.

4. Results

• $CO$ Detection: Strong absorption from $^{12}CO$ ($v=2-0$) was detected in both systems.
  • HD 110058: $\log(N_{CO}) \approx 20.0$ cm$^{-2}$, $T_{kin} \approx 133$ K.
  • HD 131488: $\log(N_{CO}) \approx 18.05$ cm$^{-2}$, $T_{kin} \approx 142$ K.
  • Results are consistent with previous UV observations (Brennan et al. 2024).
• $H_2$ Non-Detection: No statistically significant absorption was detected for the $H_2$ $v=1-0$ S(0) line at 2223.3 nm in either system.
• Column Density Limits:
  • HD 110058: $N_{H_2} < 10^{22.84}$ cm$^{-2}$.
  • HD 131488: $N_{H_2} < 10^{22.55}$ cm$^{-2}$.
• CO/$H_2$ Ratios:
  • HD 110058: CO/$H_2 > 1.35 \times 10^{-3}$.
  • HD 131488: CO/$H_2 > 3.09 \times 10^{-5}$.
• Comparison to Primordial Gas:
  • The ratio for HD 110058 is a factor of 13 higher than the canonical ISM/primordial ratio of $10^{-4}$. This definitively rules out a primordial origin for this system, indicating the gas is$H_2$-poor and likely secondary (exocometary).
  • For HD 131488, the lower limit ($>3.09 \times 10^{-5}$) formally allows for a primordial origin (as it overlaps with the lower end of primordial ratios). However, the authors argue that realistic shielding models (considering the disk's vertical thinness and lack of $H_2$) suggest CO photodissociation timescales would be shorter than the system age, further supporting a secondary origin.

5. Significance

• Resolution of the Gas Origin Debate: The study provides conclusive evidence that at least one CO-rich debris disk (HD 110058) contains gas that is compositionally distinct from primordial gas. This strongly supports the hypothesis that CO-rich belts are sustained by the outgassing of exocomets rather than being remnants of protoplanetary disks.
• Implications for Disk Evolution: The findings challenge the assumption that CO-rich belts must be primordial to explain their longevity. Instead, they suggest that secondary gas production mechanisms (cometary outgassing) are efficient enough to maintain high CO masses, even without $H_2$ shielding, provided the replenishment rate is sufficient.
• Future Directions: The paper highlights that current models of secondary gas production struggle to explain the observed CO masses without invoking unreasonably high outgassing rates or additional shielding mechanisms (like CI, which was found to be insufficient in these systems). This points to a need for more complex chemical and physical models of exocometary gas release.
• Planetary Formation Context: Understanding the volatile content (CO vs. $H_2$) of these belts is crucial for models of terrestrial planet formation, as these systems are at the age where volatile delivery to forming planets is most active.

In summary, Smith et al. (2026) utilize high-resolution IR spectroscopy to demonstrate that the gas in CO-rich debris disks is likely secondary in origin, characterized by a high CO/$H_2$ ratio that deviates significantly from primordial disk compositions.