🔭 astrophysics

Chemical Abundances of the Bioessential Elements C, O and S, and the Refractory Elements Fe and Ni, in Solar-type Exoplanet-hosting Stars from HARPS North and South

This study analyzes high-resolution HARPS spectra of 290 solar-type exoplanet-hosting stars to reveal that giant planet hosts exhibit enhanced metal abundances and lower C/O ratios compared to small planet hosts, while also identifying significant correlations between host star chemical compositions (specifically [O/H], [S/H], and [Fe/H]) and exoplanet mass that vary depending on orbital period and stellar alpha-element abundance.

Original authors: E. Costa-Almeida, L. Ghezzi, K. Cunha, V. V. Smith, J. J. Fortney

Published 2026-07-01

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: E. Costa-Almeida, L. Ghezzi, K. Cunha, V. V. Smith, J. J. Fortney

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 universe as a giant cosmic kitchen. In this kitchen, stars are the chefs, and planets are the dishes they cook up. For a long time, astronomers have been trying to figure out the recipe: Does the quality of the ingredients (the star's chemical makeup) determine what kind of dish (the planet) gets made?

This paper is a massive "taste test" of 290 solar-type stars (stars very much like our Sun) and the 373 planets orbiting them. The researchers used powerful telescopes (HARPS) to take high-resolution "photos" of the starlight, which allowed them to read the chemical ingredients list of these stars with great precision. They focused on five key elements: Carbon, Oxygen, Sulfur, Iron, and Nickel.

Here is what they found, broken down into simple concepts:

1. The "Rich Chef" Theory

The study confirms a long-held suspicion: Stars that are "rich" in metals (elements heavier than hydrogen and helium) are much more likely to have giant planets.

The Analogy: Think of making a giant cake. You need a lot of flour and sugar. If your pantry (the star) is full of these ingredients, you are likely to bake a huge cake (a giant planet like Jupiter). If your pantry is sparse, you might only bake a small cookie (a rocky planet like Earth).
The Finding: Stars hosting giant planets had higher levels of Iron, Oxygen, Sulfur, and Nickel compared to stars hosting only small planets.

2. The "Hot vs. Warm" Planet Mystery

The researchers looked at how close the planets are to their stars.

Hot Planets: These are the "sizzle" planets, orbiting very close to their star (like a hot dog on a grill).
Warm Planets: These orbit a bit further out, in a more comfortable zone.
The Finding: For small planets, the "Hot" ones tend to orbit stars that are significantly richer in metals than the stars hosting "Warm" small planets. However, for giant planets (like Jupiters), it didn't matter if they were hot or warm; their host stars were rich either way.
The Twist: The researchers noticed something strange about the metal content as you move further out from the star. They expected metal content to keep dropping as planets get further away. Instead, it dropped for a while (between 10 and 30 days of orbit) and then started rising again for planets with even longer orbits. It's like a valley in the road where the metal content dips and then climbs back up.

3. The "C/O Ratio" (The Carbon-to-Oxygen Balance)

One of the most interesting things they measured was the ratio of Carbon to Oxygen (C/O). This ratio is like the "flavor profile" of the planetary system.

The Finding: The stars hosting Jupiter-sized planets had the lowest Carbon-to-Oxygen ratios (meaning they were very Oxygen-rich).
The Surprise: The stars hosting Sub-Neptunes (planets slightly bigger than Earth but smaller than Neptune, specifically 3–4 times Earth's size) had the highest Carbon-to-Oxygen ratios.
The Analogy: It's as if the chefs making the "Sub-Neptune" dishes used a recipe heavy on carbon spices, while the chefs making the "Jupiter" dishes used a recipe heavy on oxygen.

4. The "Mass Gap" (The Missing Middle)

When they looked at the mass of the planets, they found a "gap" or a missing middle ground.

The Finding: There are lots of small, light planets and lots of huge, heavy planets, but very few planets in the middle (roughly 20 to 100 times the mass of Earth). This is often called the "Sub-Saturn Desert."
The Connection: The stars hosting the massive planets were generally richer in Oxygen and Sulfur. The stars hosting the tiny planets were right around the "average" (solar) chemical levels. This suggests that having extra Oxygen might be a key ingredient for building those massive cores that eventually swallow up gas to become giants.

5. The "Family Structure" (Single vs. Multi-Planet Systems)

The Finding: Systems with only one planet tended to have larger, more massive planets. Systems with many planets tended to have smaller, lighter planets.
The Analogy: Imagine a family. If there's only one child, they might get all the resources and grow very big. If there are many siblings, the resources are shared, and everyone stays smaller. The study found that single-planet systems are often the "big kids," while multi-planet systems are the "crowded families" of smaller worlds.

6. Comparing the Chef to the Dish

Finally, the team compared the chemical makeup of the star (the chef) to the atmosphere of the planet (the dish) for a few specific cases where we have data.

The Finding: Sometimes the planet's atmosphere has more Carbon than the star does, and sometimes less. It's not a perfect 1:1 copy.
The Conclusion: This suggests that the process of making a planet is messy and complex. The planet doesn't just copy the star's ingredients; it picks and chooses, or perhaps the ingredients change as the planet forms and moves around.

Summary

In short, this paper tells us that stars are like chefs with specific ingredient lists.

If a star has a lot of Iron, Oxygen, and Sulfur, it's likely to cook up a Giant Planet.
If a star has a high Carbon-to-Oxygen ratio, it might be more likely to cook up a Sub-Neptune.
The "recipe" changes depending on how close the planet is to the star and whether the star has a crowded family of planets or just one.

The study doesn't tell us how to build planets or find life directly, but it gives us a much clearer "cookbook" for understanding how different types of planets form in our galaxy.

Technical Summary: Chemical Abundances of Bioessential and Refractory Elements in Solar-type Exoplanet-hosting Stars

Problem and Motivation
While the correlation between stellar metallicity ([Fe/H]) and the presence of giant exoplanets is well-established, the extent of this relationship for smaller, terrestrial exoplanets remains debated. Furthermore, most abundance studies focus on iron, leaving a significant gap in homogeneous determinations of bioessential elements (C, O, S) and refractory elements (Ni) for exoplanet-hosting stars. These elements are critical for understanding protoplanetary disk chemistry, planet formation mechanisms (specifically core accretion), and planetary habitability. Previous studies have often utilized heterogeneous datasets with varying analysis methods, introducing systematic uncertainties that obscure potential star-planet connections. This work aims to provide a homogeneous, high-precision analysis of C, O, S, Fe, and Ni abundances in a large sample of solar-type stars hosting exoplanets to investigate correlations with planetary radius, orbital period, and mass.

Methodology
The authors analyzed high-resolution ( $R \sim 115,000$ ) spectra from the HARPS-North and HARPS-South spectrographs for 308 confirmed exoplanet-hosting stars (selected from the NASA Exoplanet Archive with $5000 \text{ K} \le T_{\text{eff}} \le 6500 \text{ K}$ and $\log g \ge 4.0$ ).

Stellar Parameters: Atmospheric parameters ( $T_{\text{eff}}$ , $\log g$ , [Fe/H], $v_{\text{mic}}$ ) were derived using a classical spectroscopic method based on Fe I and Fe II line excitation and ionization equilibria via an automated Python pipeline (MOOG/ARES). Evolutionary parameters (mass, radius, age) were determined using the PARAM code v1.3 with PARSEC isochrones.
Chemical Abundances: Abundances for C, O, S, and Ni were determined using the LTE code MOOG. Carbon and Nickel were derived from atomic lines (C I at 5052/5380 Å; Ni I lines). Sulfur was derived from 9 lines of the S I multiplet 8 (6743–6758 Å). Oxygen was determined using the forbidden [O I] line at 6300.304 Å, requiring careful removal of telluric contamination and blending with Ni I lines.
Exoplanet Data: Planetary radii were calculated using stellar radii from this study and literature transit depths. Planetary masses and orbital periods were taken from the NASA Exoplanet Archive.
Statistical Analysis: The study employed bootstrapping (1000 iterations) and Kolmogorov-Smirnov (K-S) tests to assess the significance of differences between populations (e.g., hot vs. warm, small vs. giant hosts). Linear regressions (OLS) were used to identify trends with planetary properties.

Key Results

Giant vs. Small Planet Hosts: Stars hosting giant exoplanets ( $R_{\text{pl}} > 4 R_{\oplus}$ ) exhibit significantly higher median [X/H] abundances (Fe, O, S, Ni) compared to stars hosting only small exoplanets. This trend holds for all elements except Carbon, where the difference is not statistically significant.
Orbital Period and Metallicity: A statistically significant anti-correlation exists between host star [Fe/H] and orbital period ( $P_{\text{orb}}$ ) for exoplanets with $P_{\text{orb}} \le 30$ days. However, contrary to a simple linear decline, [Fe/H] rises again for exoplanets with larger orbital periods ( $P_{\text{orb}} > 30$ days). This "uptrend" was confirmed by combining the sample with the California-Kepler Survey (Ghezzi et al. 2026), resulting in a homogeneous sample of 1210 exoplanets.
Hot vs. Warm Planets: Significant abundance differences between hot ( $P_{\text{orb}} \le 10$ days) and warm ( $10 < P_{\text{orb}} \le 30$ days) hosts are found for stars with only small exoplanets or those hosting at least one sub-Saturn. In contrast, stars hosting at least one Jupiter-sized planet show no significant abundance differences between hot and warm populations.
C/O Ratios: Host stars of 3–4 $R_{\oplus}$ sub-Neptunes exhibit the highest median C/O ratios ( $0.55^{+0.05}_{-0.01}$ ), while Jupiter hosts show the lowest ( $0.43^{+0.02}_{-0.03}$ ). An anti-correlation between C/O and planetary radius is observed specifically for hot exoplanets.
Mass Trends and the "Sub-Saturn Desert": A gap in the distribution of host star abundances is observed at planetary masses of $\sim 20\text{--}100 M_{\oplus}$ (the sub-Saturn desert). Stars hosting massive exoplanets generally show enhanced [O/H] and [S/H] (alpha-elements) compared to solar values, whereas low-mass hosts cluster around solar abundances. Notably, trends between [X/H] and planetary mass are robust only for warm exoplanets and for stars in the low-[ $\alpha$ /Fe] (thin disk) population; no such trends are found for hot exoplanets or high-[ $\alpha$ /Fe] (thick disk) stars.
Stellar vs. Planetary C/O: Comparisons with atmospheric C/O ratios from the literature (Changeat et al. 2022; Baburaj et al. 2025) show mixed results. While some HST/Spitzer data suggest planetary C/O > stellar C/O, higher-quality JWST/IGRINS data for specific systems show differences closer to zero or negative. The authors note a potential mass threshold ( $\sim 3.15 M_{\text{Jup}}$ ) where the relationship between stellar and planetary C/O ratios may shift, though high-quality data does not currently confirm a sharp break.

Significance and Claims
The paper claims to provide the first homogeneous determination of C, O, S, and Ni abundances for a significant number of exoplanet-hosting stars, filling gaps in existing catalogs (Vizier, Hypatia). The primary contribution is the detailed characterization of the star-planet connection for bioessential elements, revealing that:

The planet-metallicity correlation is complex and depends on both planet size and orbital period, with a non-monotonic relationship for [Fe/H] at longer periods.
The formation of giant planets appears strongly linked to oxygen and sulfur enrichment (alpha-elements), potentially highlighting the importance of the H $_2$ O ice line over CO/CO $_2$ ice lines in core accretion scenarios.
The chemical signatures of planet formation are preserved in the host star's atmosphere primarily for warm exoplanets and thin-disk stars, suggesting that hot exoplanets may have migrated from different formation environments or that their host stars' chemical signatures have been altered by dynamical processes.
The study underscores the necessity of high-quality, homogeneous abundance data to refine planet formation theories and interpret future atmospheric observations from the James Webb Space Telescope.