Carbon from Interstellar Clouds to Habitable Worlds

TL;DR

This review synthesizes how carbon is seeded from the interstellar medium into protoplanetary disks and onward into planetary bodies, emphasizing two complementary accretion pathways—pebble and planetesimal accretion—and the key disk processes that regulate carbon delivery, processing, and loss. It highlights the soot line as a critical interior boundary that governs refractory carbon survival, and it stresses the pivotal role of early pressure bumps and giant planets in modulating inward pebble flux and volatile inventories. The work links Solar System trends (e.g., C/Si gradients, NC–CC dichotomy) to broader exoplanetary diversity, arguing that Solar-like carbon architectures are not universal and that carbon-rich rocky worlds may be common in systems lacking close-in giant planets. By integrating observations from comets, meteorites, polluted white dwarfs, and JWST-era exoplanet atmospheres, the paper outlines a cohesive framework for linking disk chemistry, planet formation, and habitability across a wide range of stellar environments.

Abstract

Carbon is an essential element for a habitable world. Inner (r < 3 au) disk planetary carbon compositions are strongly influenced by supply and survival of carbonaceous solids. Here we trace the journey of carbon from the interstellar medium to the processes leading to planet formation. The review highlights the following central aspects: -Organics forming in evolved star envelopes are supplemented by aromatic molecules forming in the dense ISM to represent the seeds of (hydro)carbon supply through pervasive pebble drift to rocky planets and sub-Neptune cores. -Within the protoplanetary disk the sharp gradient in the C/Si content of Solar System bodies and mineral geochemistry outlines a tale of carbon loss from pebbles to within planetesimals and planets, and from planetary atmospheres. -Within two planet formation paradigms (pebble and planetesimal accretion) a range of planetary carbon content is possible that is strongly influenced by early (< 0.5 Myr) formation of a pressure bump that titrates drift. Overall, it is unlikely that the carbon architecture of our Solar System applies to all systems. In the absence of giant planets, carbon-rich rocky worlds and sub-Neptunes may be common. We outline observations that support their presence and discuss habitability of terrestrial worlds.

Figures (13)

• Figure 1: Schematic of some of the key processes throughout planet formation that influence carbon supply and retention. (a) Key ice/grain sublimation fronts in the disk noted (CO$_2$ between water--CO and that of Silicates at inner disk edge not shown). Planets forming in these locations will receive different compositions at birth Oberg11_C_O. Dust that grows beyond the "soot" line Lodders2004Kress2010Li21 will be rich in refractory carbon content and at greater distances molecular ices. (blow-out:) The drift of carbon-rich pebbles from the outer disk is a key process in carbon supply. Drift itself is influenced by pressure bumps that can lead to a pile-up of pebbles that can be potential sites for planetesimal and planet formation (shown on left side of disk). (b) Solids accrete ice mantles, sublimate key components at sublimation fronts and can collide leading to both growth and fragmentation. (c) Rocky planets and cores of gas-rich planets grow and gain carbon through accretion of pebbles (top) and/or planetesimals (bottom). Impacts of larger planetesimals can also strip atmospheric gas and also supply material. (d) heavy element differentiation can take carbon to the core, and the resulting magma ocean/atmosphere equilibrium can influence atmospheric composition with lighter gases escaping depending on the planet mass. Figure credit to N. Fuller and Sayo Studio.
• Figure 2: Bulk C/Si (atomic ratio) of bulk in Solar System bodies in the relative context of their potential formation location. There is a noted chemical separation of the early ($<$1 Myr) Solar System, that is associated with the formation of Jupiter Kruijer20. In this figure we denote this as "nebular chemical separation", but this has some uncertainty in location. The solar value uses the estimates obtained by 2021AA...653A.141A and 2021AA...656A.113A. The bulk Earth upper limit is taken from Li21. Bulk Silicate Earth and the C/Si ratios for NC and CC chondrites are from Bergin15. For Jupiter see § \ref{['sec:Jovian']}. The value from Hyabusa/Ryugu is given by 2022PJAB...98..227N and the value for cometary dust is an average of comets Halley and 67P Bergin15Rubin19. The C/Si ratio for Uranus and Neptune is based on the work of 2024Icar..42116217M. Error estimates for the Sun, BSE, Uranus and Neptune are smaller than the symbols. Error bars for NC and CC represent the range of values within existing samples while all other values are estimated measurement errors.
• Figure 3: Schematic of the evolution of carbonaceous grains in the diffuse to dense ISM. Carbon-rich grains form in evolved stars and are provided the diffuse interstellar medium (labeled as top down). These grains can be destroyed in shocks and processed by radiation such that only the stablest forms are provided to the dense cloud. In dense molecular clouds there is the possibility of bottom up supply of small PAHs that could lead to grain evolution subsequent phases. This figure predominantly illustrates the evolution of aromatic as opposed to aliphatic carbon carriers which is discussed in the text. Figure adapted from 2015ApJ...807...99A2021EPJD...75..152Z and kindly produced by A. Candian.
• Figure 4: (a) Signal to noise ratio of the spectral line stacked detection of the 7-ring PAH cyanocoronene in a cold dark cloud. (b) Detected molecular column density of complex carbon compounds in TMC-1. Two symbols are shown: pentagons to reflect the presence of a 5-member carbon ring and hexagons to reflect the presence of more stable six-member carbon rings. The yellow shaded area notes the regime of single ring cycles and beyond that point detected (or inferred) species contain multiple benzene and/or pentagon rings. The arrows from cyanopyrene and cyanocoronene reflect the fact that the pyrene and coronene abundances are inferred from the detections of the cyano-substituted derivative 2025NatAs...9..262W. In the molecular compounds the carbon atoms are gray, hydrogen as white, and nitrogen as blue. Figure adapted from 2025ApJ...984L..36W; the list of molecules shown in this plot is found in Table F1 of that paper. For conversion to molecular abundances the total column of molecular hydrogen along this line of sight is 1.8 $\times$10$^{22}$ cm$^{-2}$Xu_TMC_Census. Creative Commons Attribution 4.0 License.
• Figure 5: Top: Average JWST MIRI spectra of a sample of low mass T Tauri stars (0.2 M$_\odot <$ M$_* < 1.0$ M$_\odot$) probing the chemistry of the inner (r $<$ 3 au) protoplanetary disk. Bottom: Average spectra of a sample of Very Low Mass Stars, (M$_* < 0.2$ M$_\odot$). In both panels the molecular emission is modeled using a slab excitation model to match the observed emission and composition. The top spectra in each panel show the observed average spectrum and the total fit. The bottom panel provides the speciation of the emission as a function of wavelength with composition delineated via the color coding. Figure adapted from Grant25. Creative Commons Attribution 4.0 License.