Locked In Ice: how Pebble Drift and Volatile Entrapment can Significantly Impact Carbon and Oxygen Ratios in Evolving Protoplanetary Discs

TL;DR

The paper demonstrates that entrapment of CO in water ice, coupled with pebble growth and inward drift in a viscously evolving protoplanetary disc, significantly alters the gas-phase carbon and oxygen budgets. By introducing a trapped CO reservoir and a volcano-line desorption at $T\approx 130$ K, the authors show substantial increases in gas-phase C/O and C/H inside the water snowline, especially during early disc evolution when pebble flux is high. The study highlights a two-phase delivery of CO to the inner disc (via drifting pebbles, then viscous transport), with implications for planetesimal composition and planetary atmospheres, including potential relevance to Jupiter-like enrichment. These results emphasize the need to incorporate volatile entrapment into disc-models to accurately predict the volatile inventory available to forming planets and to interpret JWST observations of ice mixtures in discs.

Abstract

The complex interplay between the growth, drift, and sublimation of ice-covered pebbles can strongly influence the volatile distribution and evolution of disc composition, and therefore impact the composition of forming planets. Classic pebble drift models treat volatile species individually as sublimating at their respective snowlines, although observations from the James Webb Space Telescope (JWST) suggest that ices are likely mixed; laboratory studies suggest ice mixtures can exhibit more complex sublimation behaviours, remaining trapped beyond their nominal sublimation temperatures. We present the first model that couples pebble growth and drift with CO entrapment inside water ice - preventing a fraction (up to ~60%) of the CO from sublimating at its snowline, instead desorbing via volcanic desorption at the water crystallisation front, at 130K. Our models show that CO entrapment will significantly impact the carbon and oxygen distributions, enhancing the gas-phase C/O and C/H inside the water snowline by up to a factor of 10 over 1 Myr and a factor of a few around the CO2 snowline; O/H is also increased around the CO2 snowline, but is water-dominated in the inner disc. Entrapment therefore provides a means of introducing more carbon to the inner disc whilst retaining a large amount of water. We discuss connections to planet formation, noting that CO entrapment can increase the gas-phase heavy element content around the water snowline by up to 150%. We also consider links to JWST observations and highlight the importance of entrapment for pebble drift models to accurately model disc composition.

Figures (11)

• Figure 1: Schematic illustrating a protoplanetary disc showing a classic ice layering pebble model (top half), and a CO entrapment model (bottom half) where CO is trapped in water ice. Some of the volatile species used in this work are labelled, where the 'rocky core' consists of material with a sublimation temperature $T>300\rm{K}$. Pebbles coagulate, grow, and settle towards the midplane as they drift radially inwards. As they pass the CO snowline, the CO layer sublimates; the Classic Model loses all of its CO budget to the gas phase, whereas the Trapped CO Model retains a portion of the CO in water ice; the lower CO abundance in the Trapped CO Model is shown by a more transparent cloud of vapour. As the pebbles drift further, the water ice crystallises at $T=130\rm{K}$ and releases CO contained within the water ice in the trapped CO model via volcanic desorption; nothing occurs in the Classic Model. The water layer then sublimates at the water snowline at $T=150\rm{K}$ in the same way in both models.
• Figure 2: Surface density evolution of pure CO ice (dashed lines) and CO gas (solid lines) for our no-trapping model (blue), and also trapped CO ice (dotted lines) for trapping model (red) of parameters shown in Table \ref{['tab:model parameters']} and $\mathcal{T}_{\rm{CO}}~$=60%. Each panel shows a different snapshot in time, at $t=0$, 0.2, 1, and 3 Myr, and the top row shows $\alpha=10^{-3}$ and the bottom $10^{-4}$. CO ice can be reformed from CO condensation at its snowline, whereas trapped CO ice does not reform in our model. Water and CO snowlines are indicated by grey vertical dashed lines and are labelled above the panels in the first row; trapped CO ice is released just before the water snowline. Localised dips near snowlines in the trapped CO ice due to locally altered pebble drift speeds (see text for details). The initial conditions at $t=0$ show reduced CO gas between the water and CO snowlines in the trapping model, as 60% of the CO abundance has been converted into trapped CO ice. At $t=0.2$ Myr, pebbles deliver CO to the volcano line, which diffuses at different rates depending on $\alpha$. At $t=1$ Myr, trapped CO ice begins to sublimate into regular CO gas and provides a local boost in CO gas. By $t=3$ Myr, the gas has smeared out via diffusion.
• Figure 3: Temporal evolution of the radial gas-phase absolute carbon and oxygen ratios (left: C/O; middle: C/H; right: O/H) for our $\alpha = 10^{-4}$ model, showing both the no-trapping case (top) and $\mathcal{T}_{\rm{CO}}~$$=60\%$ trapping case (bottom). C/O=1 is shown as a grey dotted line. Each ratio begins with a static profile, shown by the dark solid line, which then evolves as pebbles grow, drift, and sublimate their ices. The trapping model most significantly alters the C/O ratio in the inner $\sim$3 au and near the CO$_2$ snowline; it also boosts the carbon content inside the CO$_2$ snowline considerably.
• Figure 4: Significance parameter $\Delta$ (see equation \ref{['eqn:significance']} in text) quantifying the impact of CO entrapment on the gas-phase absolute C/O (left), C/H (middle) and O/H (right) ratios. The $\mathcal{T}_{\rm{CO}}~$= 60% trapping model is compared against the no-trapping model for $\alpha=10^{-3}$ (top row) and $10^{-4}$ (bottom row); both models include radial drift. Higher on the colourbar (orange) indicates that the trapping model boosts the carbon and oxygen ratio compared to the fiducial, no-trapping case; lower (blue) indicates that trapping cause a reduction. For example, inside the H$_2$O snowline, the trapping model provides an enhanced C/O at $t\lesssim$ 1 Myr. The grey hatched zone outside the CO snowline indicates that these ratios are undefined due to no gas being present. The cumulative pebble flux is plotted on the left-most panel, with different colours indicating the flux through different snowlines. The biggest enhancement in C/O and C/H around the water snowline coincides with the steepest part of the cumulative pebble flux curves (i.e. when the pebble flux is highest). The volcano line is not labelled for visual clarity, but sits just outside the water snowline.
• Figure 5: Absolute C/O (top panel), C/H (middle panel) and O/H (bottom panel) ratios in the gas phase at 0.5 au. The moderate viscosity model (solid lines) quickly sweeps away CO gas released at the volcano line, leading to indistinguishable C/O evolution at $\sim$1 Myr and beyond for each trapping model. Both the high and low viscosity (dotted lines) models show considerable differences in the C/H evolution, although the low viscosity model shows larger differences when CO trapping is introduced. CO entrapment provides little variation in O/H, except for $\sim$ 5 Myr for the moderate viscosity model, when the disc is more depleted in CO for higher trapping efficiencies.