skip to main content

Title: Long-term lithium abundance signatures following planetary engulfment

Planetary engulfment events can occur while host stars are on the main sequence. The addition of rocky planetary material during engulfment will lead to refractory abundance enhancements in the host star photosphere, but the level of enrichment and its duration will depend on mixing processes that occur within the stellar interior, such as convection, diffusion, and thermohaline mixing. We examine engulfment signatures by modelling the evolution of photospheric lithium abundances. Because lithium can be burned before or after the engulfment event, it produces unique signatures that vary with time and host star type. Using mesa stellar models, we quantify the strength and duration of these signatures following the engulfment of a 1, 10, or 100 M⊕ planetary companion with bulk Earth composition, for solar-metallicity host stars with masses ranging from 0.5 to 1.4 M⊙. We find that lithium is quickly depleted via burning in low-mass host stars ($\lesssim 0.7 \, {\rm M}_\odot$) on a time-scale of a few hundred Myrs, but significant lithium enrichment signatures can last for Gyrs in G-type stars ($\sim \! 0.9 \, {\rm M}_{\odot }$). For more massive stars (1.3−1.4 M⊙), engulfment can enhance internal mixing and diffusion processes, potentially decreasing the surface lithium abundance. Our predicted signatures from exoplanet engulfment are consistent with observed lithium-rich solar-type stars and abundance enhancements in chemically inhomogeneous binary stars.

more » « less
Author(s) / Creator(s):
; ;
Publisher / Repository:
Oxford University Press
Date Published:
Journal Name:
Monthly Notices of the Royal Astronomical Society
Page Range / eLocation ID:
p. 3354-3365
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract Planetary engulfment events have long been proposed as a lithium (Li) enrichment mechanism contributing to the population of Li-rich giants ( A (Li) ≥ 1.5 dex). Using MESA stellar models and A (Li) abundance measurements obtained by the GALAH survey, we calculate the strength and observability of the surface Li enrichment signature produced by the engulfment of a hot Jupiter (HJ). We consider solar-metallicity stars in the mass range of 1–2 M ⊙ and the Li supplied by a HJ of 1.0 M J . We explore engulfment events that occur near the main-sequence turn-off (MSTO) and out to orbital separations of R ⋆ ∼ 0.1 au = 22 R ⊙ . We map our results onto the Hertzsprung–Russell Diagram, revealing the statistical significance and survival time of Li enrichment. We identify the parameter space of masses and evolutionary phases where the engulfment of a HJ can lead to Li enrichment signatures at a 5 σ confidence level and with meteoritic abundance strengths. The most compelling strengths and survival times of engulfment-derived Li enrichment are found among host stars of 1.4 M ⊙ near the MSTO. Our calculations indicate that planetary engulfment is not a viable enrichment pathway for stars that have evolved beyond the subgiant branch. For these sources, observed Li enhancements are likely to be produced by other mechanisms, such as the Cameron–Fowler process or the accretion of material from an asymptotic giant branch companion. Our results do not account for second-order effects, such as extra mixing processes, which can further dilute Li enrichment signatures. 
    more » « less

    Planet engulfment can be inferred from enhancement of refractory elements in the photosphere of the engulfing star following accretion of rocky planetary material. Such refractory enrichments are subject to stellar interior mixing processes, namely thermohaline mixing induced by an inverse mean-molecular-weight gradient between the convective envelope and radiative core. Using mesa stellar models, we quantified the strength and duration of engulfment signatures following planet engulfment. We found that thermohaline mixing dominates during the first ∼5–45 Myr post-engulfment, weakening signatures by a factor of ∼2 before giving way to depletion via gravitational settling on longer time-scales. Solar metallicity stars in the 0.5–1.2 M⊙ mass range have observable signature time-scales of ∼1 Myr–8 Gyr, depending on the engulfing star mass and amount of material engulfed. Early type stars exhibit larger initial refractory enhancements but more rapid depletion. Solar-like stars (M = 0.9–1.1 M⊙) maintain observable signatures (>0.05 dex) over time-scales of ∼20 Myr–1.7 Gyr for nominal 10 M⊕ engulfment events, with longer-lived signatures occurring for low-metallicity and/or hotter stars (1 M⊙, ∼2–3 Gyr). Engulfment events occurring well after the zero-age main sequence produce larger signals due to suppression of thermohaline mixing by gravitational settling of helium (1 M⊙, ∼1.5 Gyr). These results indicate that it may be difficult to observe engulfment signatures in solar-like stars that are several Gyr old.

    more » « less

    Dynamical evolution within planetary systems can cause planets to be engulfed by their host stars. Following engulfment, the stellar photosphere abundance pattern will reflect accretion of rocky material from planets. Multistar systems are excellent environments to search for such abundance trends because stellar companions form from the same natal gas cloud and are thus expected to share primordial chemical compositions to within 0.03–0.05 dex. Abundance measurements have occasionally yielded rocky enhancements, but a few observations targeted known planetary systems. To address this gap, we carried out a Keck-HIRES survey of 36 multistar systems, where at least one star is a known planet host. We found that only HAT-P-4 exhibits an abundance pattern suggestive of engulfment but is more likely primordial based on its large projected separation (30 000 ± 140 au) that exceeds typical turbulence scales in molecular clouds. To understand the lack of engulfment detections among our systems, we quantified the strength and duration of refractory enrichments in stellar photospheres using mesa stellar models. We found that observable signatures from 10 M⊕ engulfment events last for ∼90 Myr in 1 M⊙ stars. Signatures are largest and longest lived for 1.1–1.2 M⊙ stars, but are no longer observable ∼2 Gyr post-engulfment. This indicates that engulfment will rarely be detected in systems that are several Gyr old.

    more » « less
  4. Context. Since the discovery of exoplanetary systems, questions have been raised as to the sub-stellar companions that can survive encounters with their host star, and how this interaction may affect the internal structure and evolution of the hosting star, and particularly its surface chemical composition. Aims. We study whether the engulfment of a brown dwarf (BD) by a solar-like main-sequence (MS) star can significantly alter the structure of the star and the Li content on its surface. Methods. We performed 3D smoothed particle hydrodynamics simulations of the engulfment of a BD with masses 0.01 and 0.019 M ⊙ , on an MS star of 1 M ⊙ and solar composition, in three different scenarios: a head-on collision, a grazing collision with an impact parameter η  = 0.5  R ⊙ , and a merger. We studied the dynamics of the interaction in detail, and the relevance of the type of interaction and the mass of the BD on the final fate of the sub-stellar object and the host star in terms of mass loss of the system, angular momentum transfer, and changes in the Li abundance on the surface of the host star. Results. In all the studied scenarios, most of the BD mass is diluted in the denser region of the MS star. Only in the merger scenario a significant fraction (∼40%) of the BD material would remain in the outer layers. We find a clear increase in the surface rotational velocity of the host star after the interaction, ranging between 25 km s −1 (grazing collision) to 50 km s −1 (merger). We also find a significant mass loss from the system (in the range 10 −4  − 10 −3   M ⊙ ) due to the engulfment, which in the case of the merger may form a circumstellar disk-like structure. Assuming that neither the depth of the convective envelope of the host star nor its mass content are modified during the interaction, a small change in the surface Li abundance in the head-on and grazing collisions is found. However, in the merger we find large Li enhancements, by factors of 20 − 30, depending on the BD mass. Some of these features could be detected observationally in the host star, provided they remained for a long enough time. Conclusions. In our 3D simulations, a sizable fraction of the BD survives long enough to be mixed with the inner core of the MS star. This is at odds with previous suggestions based on 1D simulations. In some cases the final surface rotational velocity is very high, coupled with enough mass loss that may form a circumstellar disk. Merger scenarios tend to dilute considerably more BD material on the surface of the MS star, which could be detected as a Li-enhancement. The dynamic of the simulated scenarios suggests the development of asymmetries in the structure of the host star that can only be tackled with 3D codes, including the long-term evolution of the system. 
    more » « less
  5. Abstract Type Ia supernovae are critical for feedback and elemental enrichment in galaxies. Recent surveys like the All-Sky Automated Survey for Supernova (ASAS-SN) and the Dark Energy Survey (DES) find that the specific supernova Ia rate at z ∼ 0 may be ≲ 20 − 50 × higher in lower-mass galaxies than at Milky Way-mass. Independently, observations show that the close-binary fraction of solar-type Milky Way stars is higher at lower metallicity. Motivated by these observations, we use the FIRE-2 cosmological zoom-in simulations to explore the impact of metallicity-dependent rate models on galaxies of $M_* \sim 10^7\, \rm {M}_{\odot }-10^{11}\, \rm {M}_{\odot }$. First, we benchmark our simulated star-formation histories (SFHs) against observations, and show that assumed stellar mass functions play a major role in determining the degree of tension between observations and metallicity-independent rate models, potentially causing ASAS-SN and DES observations to agree more than might appear. Models in which the supernova Ia rate increases with decreasing metallicity ($\propto Z^{-0.5 \; \rm {to} \; -1}$) provide significantly better agreement with observations. Encouragingly, these rate increases (≳ 10 × in low-mass galaxies) do not significantly impact galaxy masses and morphologies, which remain largely unaffected except for our most extreme models. We explore implications for both [Fe/H] and [$\alpha /\rm {Fe}$] enrichment; metallicity-dependent rate models can improve agreement with the observed stellar mass-metallicity relations in low-mass galaxies. Our results demonstrate that a range of metallicity-dependent rate models are viable for galaxy formation and motivate future work. 
    more » « less