Post-formation Origin of Low Atmospheric C/O Ratios in Giant Exoplanets and the Role of Internal Heating

Recent analyses of JWST transmission spectroscopy have suggested that some transiting giant gas planets and gas dwarfs have low atmospheric C/O ratios, comparable to or lower than the solar value (C/O_solar ~= 0.55), as well as high intrinsic temperatures reaching T_int ≥ 300 K (e.g., Sing et al. 2024; Welbanks et al. 2024). These inferred T_int values are higher than the T_int ~ 100 K predicted by conventional cooling models (Fortney et al. 2020), and are difficult to explain by standard cooling evolution alone.

The origin of low atmospheric C/O ratios may be attributed either to planet formation processes or to post-formation internal evolution. As a planet formation origin, oxygen-rich heavy elements may be supplied by the evaporation of pebbles that drift radially in the protoplanetary disk, or by the evaporation of pebbles accreted into the planetary envelope during formation (e.g., Brouwers et al. 2018). In this study, we instead focus on an internal-evolution origin, namely the transport of heavy elements from a diluted core to the outer envelope.

Observations of Jupiter and Saturn by Juno and Cassini suggest the existence of diluted cores, in which the heavy-element abundance decreases continuously from the center outward, rather than being confined to a compact central core (e.g., Wahl et al. 2017; Mankovich & Fuller 2021). In planets with such diluted cores, heavy elements distributed in the interior can be transported outward over time, thereby altering the atmospheric composition after formation (Knierim & Helled 2025). When this process is applied, even planets that initially accreted high-C/O gas, as predicted by the static disk chemistry model (Öberg et al. 2011), can reduce their atmospheric C/O ratios to solar or sub-solar values if oxygen-rich heavy elements from the diluted core are transported into the atmosphere after formation.

On the other hand, realizing the observed T_int ≥ 300 K may require an additional heat source. One possible mechanism is internal heating, in which a fraction of the incident stellar irradiation is transported into the planetary interior (Komacek & Youdin 2017). Komacek & Youdin (2017) showed that internal heating can delay planetary cooling and contraction by suppressing the vigorous convection that occurs in gas planets. However, because the transport of heavy elements from the diluted core to the outer envelope is mainly driven by convective mixing, suppression of convection by internal heating may inhibit the post-formation decrease in atmospheric C/O. Thus, internal heating that maintains a high T_int and diluted-core mixing that produces a low atmospheric C/O ratio may act as competing processes.

In this study, we introduce internal heating into giant gas planet models with diluted cores and investigate the conditions under which C/O ≤ 0.55 and T_int ≥ 300 K can be simultaneously achieved through post-formation internal evolution. We use the stellar evolution code MESA (Paxton et al. 2011–2019; Jermyn et al. 2023) and the planetary module MESPA (Helled, Müller, & Knierim 2025). As the initial internal structure, we adopt a Jupiter-mass diluted-core model based on Knierim & Helled (2025), and set the initial atmospheric C/O ratio to the high gas-phase value predicted by the disk chemistry model of Öberg et al. (2011). Following Komacek & Youdin (2017), we parameterize internal heating by the heating efficiency and the heating depth. The heating efficiency is defined as the fraction of the incident irradiation deposited as heat in the planetary interior, and is treated here as a free parameter controlling the strength of internal heating. The heating depth specifies where this energy is deposited. We use a model with T_eq = 1000 K as a baseline. For a given heating depth, the deposited heating power depends on the product of the incident irradiation and the heating efficiency, so the heating efficiency controls the effective strength of internal heating in our baseline models.

We find that, for the T_eq = 1000 K baseline model, a heating efficiency of 1% can maintain T_int > 300 K while reducing the atmospheric C/O ratio to a sub-solar value through the transport of oxygen-rich heavy elements from the diluted core. This 1% case corresponds to a representative heating efficiency discussed by Komacek & Youdin (2017) for explaining the radius inflation of hot Jupiters. In contrast, when the heating efficiency is increased to 10% and the deposited heat reaches the convective envelope, a higher T_int can be achieved, but a radiative layer forms within the envelope and inhibits the transport of heavy elements from the diluted core to the outer envelope. As a result, the post-formation decrease in C/O becomes inefficient. The 10% case corresponds to the upper limit of the heating efficiency adopted by Komacek & Youdin (2017), and is used here as a strong-heating comparison model.

These results have important implications for interpreting the origin of low atmospheric C/O ratios. When internal heating is relatively weak, or when the deposited heat does not strongly affect the convective envelope, convective transport from the diluted core can remain efficient. In such planets, low atmospheric C/O ratios may not directly reflect the composition acquired during formation; even planets that initially accreted high-C/O gas can acquire solar or sub-solar atmospheric C/O ratios after formation through the upward transport of oxygen-rich heavy elements from the diluted core. In contrast, when internal heating is strong and reaches the convective envelope, a radiative layer can form within the planet and suppress convective heavy-element transport, making post-formation C/O reduction inefficient. Because the deposited heating power increases with incident irradiation, strongly irradiated planets can experience strong internal heating even with a modest heating efficiency. If such planets nevertheless show low C/O ratios, their atmospheric compositions may more directly reflect the composition acquired during formation rather than subsequent internal evolution. Therefore, strongly irradiated planets are promising targets for distinguishing the origin of low atmospheric C/O ratios.

In this presentation, we discuss how internal heating affects the evolution of T_int and atmospheric C/O ratios in giant planets with diluted cores, and explore the possibility of distinguishing the origin of atmospheric composition.