No Arabic abstract
Between 25-50 % of white dwarfs (WD) present atmospheric pollution by metals, mainly by rocky material, which has been detected as gas/dust discs, or in the form of photometric transits in some WDs. Planets might be responsible for scattering minor bodies that can reach stargazing orbits, where the tidal forces of the WD can disrupt them and enhance the chances of debris to fall onto the WD surface. The planet-planet scattering process can be triggered by the stellar mass-loss during the post main-sequence evolution of planetary systems. In this work, we continue the exploration of the dynamical instabilities that can lead to WD pollution. In a previous work we explored two-planet systems found around main-sequence (MS) stars and here we extend the study to three-planet system architectures. We evolved 135 detected three-planet systems orbiting MS stars to the WD phase by scaling their orbital architectures in a way that their dynamical properties are preserved by using the $N$-body integrator package Mercury. We find that 100 simulations (8.6 %) are dynamically active (having planet losses, orbit crossing and scattering) on the WD phase, where low mass planets (1-100 $mathrm{M}_oplus$) tend to have instabilities in Gyr timescales while high mass planets ($>100~mathrm{M}_oplus$) decrease the dynamical events more rapidly as the WD ages. Besides, 19 simulations (1.6 %) were found to have planets crossing the Roche radius of the WD, where 9 of them had planet-star collisions. Our three-planet simulations have an slight increase percentage of simulations that may contribute to the WD pollution than the previous study involving two-planet systems and have shown that planet-planet scattering is responsible of sending planets close to the WD, where they may collide directly to the WD, become tidally disrupted or circularize their orbits, hence producing pollution on the WD atmosphere.
Asteroid material is detected in white dwarfs (WDs) as atmospheric pollution by metals, in the form of gas/dust discs, or in photometric transits. Within the current paradigm, minor bodies need to be scattered, most likely by planets, into highly eccentric orbits where the material gets disrupted by tidal forces and then accreted onto the star. This can occur through a planet-planet scattering process triggered by the stellar mass loss during the post main-sequence evolution of planetary systems. So far, studies of the $N$-body dynamics of this process have used artificial planetary system architectures built ad hoc. In this work, we attempt to go a step further and study the dynamical instability provided by more restrictive systems, that, at the same time allow us an exploration of a wider parameter space: the hundreds of multiple planetary systems found around main-sequence (MS) stars. We find that most of our simulated systems remain stable during the MS, Red and Asymptotic Giant Branch and for several Gyr into the WD phases of the host star. Overall, only $approx$ 2.3$%$ of the simulated systems lose a planet on the WD as a result of dynamical instability. If the instabilities take place during the WD phase most of them result in planet ejections with just 5 planetary configurations ending as a collision of a planet with the WD. Finally 3.2$%$ of the simulated systems experience some form of orbital scattering or orbit crossing that could contribute to the pollution at a sustained rate if planetesimals are present in the same system.
The atmospheres of between one quarter and one half of observed single white dwarfs in the Milky Way contain heavy element pollution from planetary debris. The pollution observed in white dwarfs in binary star systems is, however, less clear, because companion star winds can generate a stream of matter which is accreted by the white dwarf. Here we (i) discuss the necessity or lack thereof of a major planet in order to pollute a white dwarf with orbiting minor planets in both single and binary systems, and (ii) determine the critical binary separation beyond which the accretion source is from a planetary system. We hence obtain user-friendly functions relating this distance to the masses and radii of both stars, the companion wind, and the accretion rate onto the white dwarf, for a wide variety of published accretion prescriptions. We find that for the majority of white dwarfs in known binaries, if pollution is detected, then that pollution should originate from planetary material.
Metal pollution in white dwarf atmospheres is likely to be a signature of remnant planetary systems. Most explanations for this pollution predict a sharp decrease in the number of polluted systems with white dwarf cooling age. Observations do not confirm this trend, and metal pollution in old (1-5 Gyr) white dwarfs is difficult to explain. We propose an alternative, time-independent mechanism to produce the white dwarf pollution. The orbit of a wide binary companion can be perturbed by Galactic tides, approaching close to the primary star for the first time after billions of years of evolution on the white dwarf branch. We show that such a close approach perturbs a planetary system orbiting the white dwarf, scattering planetesimals onto star-grazing orbits, in a manner that could pollute the white dwarfs atmosphere. Our estimates find that this mechanism is likely to contribute to metal pollution, alongside other mechanisms, in up to a few percent of an observed sample of white dwarfs with wide binary companions, independent of white dwarf age. This age independence is the key difference between this wide binary mechanism and others mechanisms suggested in the literature to explain white dwarf pollution. Current observational samples are not large enough to assess whether this mechanism makes a significant contribution to the population of polluted white dwarfs, for which better constraints on the wide binary population are required, such as those that will be obtained in the near future with Gaia.
Circumstellar disks of planetary debris are now known or suspected to closely orbit hundreds of white dwarf stars. To date, both data and theory support disks that are entirely contained within the preceding giant stellar radii, and hence must have been produced during the white dwarf phase. This picture is strengthened by the signature of material falling onto the pristine stellar surfaces; disks are always detected together with atmospheric heavy elements. The physical link between this debris and the white dwarf host abundances enables unique insight into the bulk chemistry of extrasolar planetary systems via their remnants. This review summarizes the body of evidence supporting dynamically active planetary systems at a large fraction of all white dwarfs, the remnants of first generation, main-sequence planetary systems, and hence provide insight into initial conditions as well as long-term dynamics and evolution.
WD J204713.76-125908.9 is a new addition to the small class of white dwarfs with helium-dominated photospheres that exhibit strong Balmer absorption lines and atmospheric metal pollution. The exceptional abundances of hydrogen observed in these stars may be the result of accretion of water-rich rocky bodies. We obtained far-ultraviolet and optical spectroscopy of WD J204713.76-125908.9 using the Cosmic Origin Spectrograph on-board the Hubble Space Telescope and X-shooter on the Very Large Telescope, and identify photospheric absorption lines of nine metals: C, O, Mg, Si, P, S, Ca, Fe and Ni. The abundance ratios are consistent with the steady state accretion of exo-planetesimal debris rich in the volatile elements carbon and oxygen, and the transitional element sulphur, by factors of seventeen, two, and four respectively compared to bulk Earth. The parent body has a composition akin to Solar System carbonaceous chondrites, and the inferred minimum mass, $1.6 times 10^{20}$ g, is comparable to an asteroid 23 km in radius. We model the composition of the disrupted parent body, finding from our simulations a median water mass fraction of eight per cent.