Hot Jupiters Aren’t As Lonely As We Thought


The Friends of Hot Jupiters (FOHJ) project is a systematic search for planetary- and stellar-mass companions in systems that have known hot Jupiters — short-period, gas-giant planets. This survey has discovered that many more hot Jupiters may have companions than originally believed.

Missing Friends

FOHJ was begun with the goal of better understanding the systems that host hot Jupiters, in order to settle several longstanding issues.

The first problem was one of observational statistics. We know that roughly half of the Sun-like stars nearby are in binary systems, yet we’ve only discovered a handful of hot Jupiters around binaries. Are binary systems less likely to host hot Jupiters? Or have we just missed the binary companions in the hot-Jupiter-hosting systems we’ve seen so far?

An additional issue relates to formation mechanisms. Hot Jupiters probably migrated inward from where they formed out beyond the ice lines in protoplanetary disks — but how?

AO image

This median-stacked image, obtained with adaptive optics, shows one of the newly-discovered stellar companions to a star hosting a hot Jupiter. The projected separation is ~180 AU. [Ngo et al. 2015]

Observations reveal two populations of hot Jupiters: those with circular orbits aligned with their hosts’ spins, and those with eccentric, misaligned orbits. The former population support a migration model dominated by local planet-disk interactions, whereas the latter population suggest the hot Jupiters migrated through dynamical interactions with distant companions. A careful determination of the companion rate in hot-Jupiter-hosting systems could help establish the ability of these two models to explain the observed populations.

Search for Companions

The FOHJ project began in 2012 and studied 51 systems hosting known, transiting hot Jupiters — with roughly half on circular, aligned orbits and half on eccentric, misaligned orbits. The survey consisted of three different, complementary components:

  • Study 1
    Lead author: Heather Knutson (Caltech)
    Technique: Long-term radial velocity monitoring
    Searching for: Planetary companions at 1–20 AU from the star
  • Study 2
    Lead author: Henry Ngo (Caltech)
    Technique: Adaptive-optics imaging
    Searching for: Stellar companions at 50–2000 AU from the star
  • Study 3
    Lead author: Danielle Piskorz (Caltech)
    Technique: Spectroscopy
    Searching for: Any additional stellar companions at <125 AU from the star
binary fractions

The companion fraction found within Study 2, the adaptive-optics imagine search. The three curves show the total, the systems with hot Jupiters on aligned and circular orbits, and those with hot Jupiters on misaligned and eccentric orbits. [Ngo et al. 2015]

Migration Implications

Using these three different techniques, the team found a significant number of both planetary and stellar companions that had not been previously detected. After correcting their results for completeness, they found a multiple-star rate of ~50% for these systems, resolving the problem of the missing companions. So really, we just weren’t looking hard enough for the companions previously.

Intriguingly, the binary companion rate found for these hot Jupiter systems is higher than the average rate for the field stars (which is below 25% for the semimajor-axis range the FOHJ studies are sensitive to). This suggests that companion stars may indeed play a role in hot Jupiter formation and migration.

That said, none of the three studies found a significant difference in the binary fraction for aligned versus misaligned hot Jupiters — which means that the answer is not as simple as thought, with companion stars causing the misaligned planets. Thus, while hot Jupiters’ “friends” may play a role in their formation and migration, we still have work to do in understanding what that role is.


Danielle Piskorz et al 2015 ApJ 814 148. doi:10.1088/0004-637X/814/2/148
Henry Ngo et al 2015 ApJ 800 138. doi:10.1088/0004-637X/800/2/138
Heather A. Knutson et al 2014 ApJ 785 126. doi:10.1088/0004-637X/785/2/126