Table of Contents
Fetching ...

Stars stably accreting from substellar objects

Aaron Householder, Kaitlyn Shin, Kevin B. Burdge, Thomas R. Marsh, Saul A. Rappaport, Kareem El-Badry, Joheen Chakraborty, Emma Chickles, Fei Dai, Matthew J. Graham, S. R. Kulkarni, Pablo Rodríguez-Gil, Andrew Vanderburg, Samuel Whitebook

Abstract

Substellar objects such as brown dwarfs and planets are generally expected to remain detached from their main-sequence host stars unless orbital decay or stellar expansion brings them into contact, leading to rapid engulfment and destruction. Such a fate is predicted for the Earth and other rocky planets in our solar system; however, in certain cases, theory also allows for stable long-lived mass transfer from a substellar object onto its main-sequence host, though such accretion has never been directly observed. Here we report the first direct observations of stable mass transfer from a substellar object onto a main-sequence star. In particular, we identify two binaries, ZTF J0440+2325 and ZTF J1444+4820, with orbital periods of just 87 and 67 minutes, respectively, in which a brown dwarf stably transfers mass onto an M dwarf companion. These systems demonstrate that the fate of some substellar objects is not rapid engulfment and destruction, but instead gradual consumption for potentially billions of years.

Stars stably accreting from substellar objects

Abstract

Substellar objects such as brown dwarfs and planets are generally expected to remain detached from their main-sequence host stars unless orbital decay or stellar expansion brings them into contact, leading to rapid engulfment and destruction. Such a fate is predicted for the Earth and other rocky planets in our solar system; however, in certain cases, theory also allows for stable long-lived mass transfer from a substellar object onto its main-sequence host, though such accretion has never been directly observed. Here we report the first direct observations of stable mass transfer from a substellar object onto a main-sequence star. In particular, we identify two binaries, ZTF J0440+2325 and ZTF J1444+4820, with orbital periods of just 87 and 67 minutes, respectively, in which a brown dwarf stably transfers mass onto an M dwarf companion. These systems demonstrate that the fate of some substellar objects is not rapid engulfment and destruction, but instead gradual consumption for potentially billions of years.
Paper Structure (7 sections, 5 equations, 10 figures, 1 table)

This paper contains 7 sections, 5 equations, 10 figures, 1 table.

Figures (10)

  • Figure 1: Photometric and spectroscopic observations of ZTF J0440+2325. (a) Phase-folded HiPERCAM light curves in the $u$, $g$, $r$, $i$, and $z$ bands, showing large-amplitude chromatic variability produced by an accretion hot spot on the M dwarf. (b) Keck/LRIS blue-arm spectrum obtained near peak brightness ($\phi = 0.43$), with transient Balmer absorption arising from the accretion hot spot. (c) Keck/LRIS red-arm spectrum obtained near minimum brightness ($\phi = 0.91$), showing molecular absorption features consistent with an M8 dwarf template (red). (d) Radial-velocity measurements of the M dwarf from the ESI spectra (blue) and LRIS spectra (red) as a function of orbital phase. The best-fitting radial velocity model has a semi-amplitude of $K = 24.7 \pm 2.6\,\mathrm{km\,s^{-1}}$, demonstrating that the unseen companion lies in the substellar mass regime.
  • Figure 2: Photometric and spectroscopic observations of ZTF J1444+4820. (a) Phase-folded HiPERCAM light curves of ZTF J1444+4820 in the $u$, $g$, $r$, $i$, and $z$ bands, showing strong wavelength-dependent variability from an accretion hot spot on the M dwarf that is eclipsed by the brown dwarf companion near peak brightness. (b) Phase-resolved Keck/LRIS blue-arm spectra obtained over a portion of the orbit, with transient Balmer absorption appearing near peak brightness when the hot spot dominates the optical flux. (c) SPHEREx near-infrared spectro-photometry covering $0.75$--$3.75\,\mu\mathrm{m}$ (red points with error bars) compared to a BT--Settl atmosphere model with $T_{\mathrm{eff}} = 2700\,\mathrm{K}$ (black line), indicating that the dominant photospheric component is indeed a late-type M dwarf.
  • Figure 3: Ballistic stream trajectories of test particles in the co-rotating frame of ZTF J0440+2325. The black points mark the centers of mass of the two components and the solid black circle shows the surface of the accretor. The dashed black curves indicate Roche-lobe equipotential surfaces, with the saddle point marking the inner Lagrange point (L1). The red curve shows trajectories of test particles launched from near L1, which directly impact the accretor surface rather than forming an accretion disk. The impact point (IP) is offset from the substellar point (SP) on the accretor by $\lambda_{\mathrm{IP}}-\lambda_{\mathrm{SP}} = 14.8^\circ$, corresponding to a phase offset $\Delta\phi \simeq 0.04$, consistent with the observed phase of maximum brightness in the HiPERCAM data.
  • Figure 4: Light-curve model fits for ZTF J0440+2325. HiPERCAM $g$-band photometry (green) compared to two different hot-spot models computed with lcurve (black and purple). The residuals (data minus model) are shown below the fits. Both models reproduce the overall variability but do not fully capture the detailed light-curve structure, indicating limitations of simplified spot prescriptions in reproducing the observed hot-spot structure.
  • Figure 5: Light-curve model fits for ZTF J1444+4820. HiPERCAM $g$-band light curve of ZTF J1444+4820 (green points with error bars) compared to two lcurve models (black and purple curves). The lower panel shows the residuals (data minus model) for each fit.
  • ...and 5 more figures