X-ray binaries

Most stars are believed to be born as couples (contrary to the Sun) but when one of them runs out of fuel, pressure is no longer able to counteract the collapsing tendency of gravity and it turns into a compact object : a white dwarf, a neutron star or a black hole. Those tiny objects are usually tough to detect but as part of a binary system, they can interact with their surviving companion star in a different way than during their shining stellar phase. Depending on the properties of the binary system, the surroundings of the compact object can be either a bed of roses or a wild stormy flow.
In the late 50’s-early 60’s, as the space race was raging, astronomers designed new instruments, able to watch the sky in a color range far beyond the scope of our eyes called X-rays. In 1962, a team led by Riccardo Giacconi detected the first extrasolar X-ray source, Scorpius X-1. Since no star is expected to emit at such wavelength, the nature of this object remained unclear until Iosif Shklovsky suggested in 67 that it could come from a « stream of gas » flowing « from a secondary component of a close binary system toward […] a neutron star ». The X-ray binaries family was born.

Roche lobe overflow

I first exploited the deformation induced by the Roche potential to disentangle ellipsoidal light variations, Doppler boosting and mutual illumination in close stellar binary systems. Yet, it is only in my thesis work that I started to study Roche lobe overflowing systems in an X-ray binary context. Much progress have been done for the last 20 years regarding accretion discs dynamics but its conditions of formation remain to be explored. The important variability of those systems, overwhelmed with violent outbursts and spectral reconfigurations[A], suggests that the external feeding of the disc, at the orbital scale, might determine some properties of the disc and modulate its periodicities and duty cycles.

Wind accretion in HMXB

Some X-ray binaries host massive stars underflowing their Roche lobe, which brings up the question of the fueling process of the disc if any. Since massive stars undergo important outflows at permanent rates which can reach 10-6 solar masses per year[B], a fraction of this wind captured bya compact object might be enough to account for the levels of X-ray emission we detect. Though, the low levels of angular momentum carried by those flows would make the formation of a wide spread disc more uncertain than in the Roche lobe overflow sketch.
Such accretion flows have been firstly suggested in the late 30's by Hoyle & Lyttleton before being refined by Bondi & Hoyle in the mid-40's. When numerical simulations confronted this models, they pinpointed discrepancies and instabilities. Several theoretical studies were undergone to further characterize those axisymmetric flows and bridge the gap between 2.5D ballistic and 1D hydrodynamical approaches[C]. With Fabien, we designed a numerical setup to settle down the issues concerning the dependency of the results on the size of the accretor and to assess the new theoretical predictions[D]. This setup turned out to be a robust basis to work with given its degree of consistency with the predicted properties of the flow. Gathering the accretion scale and the size of the accretor within one reliable simulation enabled us to confidently add the non axisymmetric components entailed by the orbital angular momentum of X-ray binary systems.

Short-period exoplanets

Since the first exoplanet orbiting a main-sequence star was discovered in 1995 by Mayor and Queloz[E], the number of confirmed exoplanets has jumped to almost two thousands. Initially, the instrumental biases of the detection methods could not discard the possibility that the objects spotted were merely marginal cases : hot Jupiters, eccentric or misaligned orbits... nothing like the well-behaved planets of the Solar system. However, with devoted instruments like the Kepler & Corot satellites around 2010, it turned out that outliers were not that rare, if not the rule.
As we were investigating short-periodicities in Kepler's first quarters released in 2011, we ran into a 16 hours signal whose amplitude was highly varying from a transit to the next one. As deep as a one percent of the stellar luminosity, the dimming could also vanish below the .2% noise level. We designed several scenarii like a peanut-planet one I appraised both numerically and theoretically but the most reliable we agreed on was a comet-like tail absorbing feature, later confirmed by short cadence data from a follow-up with Kepler. Indeed, whatever the nature of the orbiting body, its temperature is likely to reach the boiling point of some silicates-elements so if the planet turns out to be rocky and small enough (to get a low gravitational field), dust could be lifted up from the surface and account for the continuous absorption in the tail. Exoplanets undergoing gas escape had already been observed[F], so as massive lava planets[G] but it was the first time whole bunches of dust leaving a planet were observed : the tiny body was literally abraded by the merciless flux from its K-type host star. Further investigations pinpointed the role of the stellar activity cycle[H] and the characteristic size of the grains in the obscuring cloud[I][J]. Since then, up to three disintegrating planets could have been identified[K].


A. Remillard, R. A. and McClintock, J. E. (2006), ‘X-Ray Properties of Black- Hole Binaries’, 44, 49–92.
B. Vink, J. S. (2000), Radiation-driven Wind Models of Massive Stars, PhD thesis, Universiteit Utrecht
C. Foglizzo, T. and Ruffert, M. (1996), Astron. Astrophys. 361, 22.
D. El Mellah, I., Casse, F. (2015)
E. Mayor, M. and Queloz, D. (1995), ‘A Jupiter-mass companion to a solar-type star’, 378, 355–359.
F. Charbonneau, D., Brown, T. M., Latham, D. W. and Mayor, M. (2000), ‘Detection of Planetary Transits Across a Sun-like Star’, 529, L45–L48.
G. Léger, A., Rouan, D., Schneider, J., Barge, P., Fridlund, M., Samuel, B., Ollivier, M., Guenther, E., Deleuil, M., Deeg, H. J. et al (2009), ‘Transiting exoplanets from the CoRoT space mission. VIII. CoRoT-7b : the first super-Earth with measured radius’, 506, 287–302.
H. Croll, B., Rappaport, S. and Levine, A. M. (2015), ‘The relation between the transit depths of KIC 12557548b and the stellar rotation period’, 449, 1408–1421.
I. Croll, B., Rappaport, S., DeVore, J., Gilliland, R. L., Crepp, J. R., Howard, A. W., Star, K. M., Chiang, E., Levine, A. M., Jenkins, J. M., Albert, L., Bonomo, A. S., Fortney, J. J. and Isaacson, H. (2014), ‘Multiwave- length Observations of the Candidate Disintegrating Sub-Mercury KIC 12557548b’, 786, 100.
J. Bochinski, J. J., Haswell, C. A., Marsh, T. R., Dhillon, V. S. and Little- fair, S. P. (2015), ‘Direct Evidence for an Evolving Dust Cloud from the Exoplanet KIC 12557548 b’, 800, L21.
K. Sanchis-Ojeda, R., Rappaport, S., Pallé, E., Delrez, L., DeVore, J., Gandolfi, D., Fukui, A., Ribas, I., Stassun, K. G. et al (2015), 'The K2- ESPRINT Project I : Discovery of the Disintegrating Rocky Planet with a Cometary Head and Tail EPIC 201637175b', ArXiv e-prints.