Atmospheric characterization of HIP 67522 b with VLT/CRIRES+. VLT/CRIRES+ suggests a heavier planet and hints at deuterium fractionation

Atmospheric characterization of HIP 67522 b with VLT/CRIRES+. VLT/CRIRES+ suggests a heavier planet and hints at deuterium fractionation
Notice: This research summary and analysis were automatically generated using AI technology. For absolute accuracy, please refer to the [Original Paper Viewer] below or the Original ArXiv Source.

Young transiting exoplanets provide unique opportunities to probe planetary atmospheres during the critical early phases of evolution. HIP67522b, a 17Myr old hot Jupiter with an extraordinarily low bulk density, represents an ideal target for high-resolution transmission spectroscopy. We aim to characterize the atmospheric composition, thermal structure, and dynamics of HIP67522b using ground-based high-resolution near-infrared spectroscopy. We obtained high-resolution spectra with VLT/CRIRES+ in the K2166 band during a transit on 30 January 2025. We applied cross-correlation techniques and Bayesian nested sampling retrievals to constrain molecular abundances, temperature structure, and atmospheric dynamics. We detect H$_2$O at 20$σ$ and CO at 5$σ$, confirming the extremely extended atmosphere of this low-mass giant. A velocity offset of $-2.9 \pm 0.2$kms$^{-1}$ indicates day-to-night winds. The rotation velocity is constrained to $<1.8$kms$^{-1}$ at 3$σ$, consistent with tidal locking. Retrieval analysis suggests a planetary mass of 29.8 $\pm$ 3 Earth masses and a vertically isothermal atmosphere. This mass is two times larger than the mass estimated from JWST atmospheric observations and inconsistent at 3$σ$ hence leaving a doubt on the actual planetary density of the planet. Using the mass derived derived from the CRIRES+ data, we derive a C/O ratio of $0.83 \pm 0.09$, about 1.5 times solar, and a subsolar metallicity [C+O/H]$= -0.8 \pm 0.4$ which can be increased if the atmosphere is cloudy, a degeneracy our data alone cannot resolve. We report a tentative 2$σ$ detection of HDO with an extreme enrichment factor of $\sim$1000 relative to the protosolar D/H ratio. If confirmed, this would be the first detection of deuterium in an exoplanet atmosphere and would require intense escape rate to be explained.


💡 Research Summary

HIP 67522b is a 17‑million‑year‑old hot Jupiter orbiting a young G‑type star in the Sco‑Cen association. Its unusually large radius (~10 R⊕) and extremely low bulk density (<0.1 g cm⁻³) make it an ideal target for high‑resolution transmission spectroscopy, as its atmospheric scale height is comparable to the planetary radius.

The authors obtained 92 high‑resolution (R≈100 000) near‑infrared spectra with VLT/CRIRES+ in the K‑band (centered at 2166 nm) during a full transit on 30 January 2025. After standard calibrations, they applied the ATMOSPHERIX pipeline, which includes alignment in the stellar rest frame, telluric correction, continuum normalization, variance‑based outlier masking, and a principal‑component analysis (removing six components) to suppress correlated noise.

Cross‑correlation of the reduced data with isothermal model spectra generated by petitRADTRANS revealed a 20‑σ detection of water (H₂O) and a 5‑σ detection of carbon monoxide (CO). The peak correlation occurs at a velocity offset of –3 km s⁻¹ relative to the systemic velocity, indicating strong day‑to‑night winds on the planetary limb. No other molecules (e.g., CH₄, NH₃, HCN, SO₂) were detected.

To retrieve atmospheric parameters, the team performed Bayesian nested‑sampling using PyMultiNest. They modeled the atmosphere as isothermal, fitting gravity (g), temperature (T), individual molecular mixing ratios, and a rotational broadening term. Planetary mass was inferred indirectly from gravity via the relation M = g R² T/(G R_TESS), where R_TESS = 0.892 R_J is the radius measured by TESS. The combined likelihood includes a TESS transit depth term and a high‑resolution spectroscopy term with a scaling factor to account for model amplitude.

The retrieval yields a planetary mass of 29.8 ± 3 M⊕, roughly twice the mass (13.8 ± 1.0 M⊕) previously derived from JWST/NIRSpec transmission spectra. This discrepancy is statistically significant at the 3‑σ level, suggesting either systematic differences between low‑ and high‑resolution techniques or unmodelled atmospheric effects (e.g., clouds, non‑isothermal structure).

Chemical composition results show a carbon‑to‑oxygen ratio of C/O = 0.83 ± 0.09, about 1.5 times the solar value, and a sub‑solar metallicity of


Comments & Academic Discussion

Loading comments...

Leave a Comment