The 51 Pegasi b, discovered by Didier Queloz and Michel Mayor is classified as a hot Jupiter and not considered to be a star. I had read a bit about the minimum mass necessary for being a brown dwarf at some 13 Jovian masses, also the following question "Can Jupiter be ignited?" which has an answer stating that clear boundaries have not been drawn regarding gas giants and brown dwarfs. My question is how was it concluded that a Hot Jupiter different is from star types like brown dwarfs, in the sense that they could assert it wasn't a star they were observing ( keeping in mind the possibility of more massive ones like WASP-18-B (around 10 Jovian masses)).

Specifically how does one eliminate the possibility that it could have been a star at the end of its life-cycle for we could have considered it to be a smaller star of a binary system (considering the case of distance to be pretty close to the central star) ?

Apologies if the question sounds too naive.


1 Answer 1


Your question is not naive at all, don't worry.

51 Peg b has been found via the radial velocity method. Upon detection, this method automatically gives a minimum mass of the companion, which is $M_{\rm min} = M \; \sin(i)$, where $M$ is the unknown actual planetary mass and $i$ is the (generally) unknown inclination of the orbit towards the observer.

It was determined that $M_{\rm min}\approx 0.5 M_{\rm Jup}$, and for isotropically distributed orientations of orbits on the sky, the expectation value of $\sin(i)$ is about unity. Extremal values of $i$ that would make $\sin(i) \ll 1$ and thus $M \gg M_{\rm Jup}$, are improbable, but not excluded. In their paper, Mayor & Queloz cite a $1\%$ chance that the planet is above $4 M_{\rm Jup}$ and $1/40.000$ that it is above the deuterium-burning limit of $13 M_{\rm Jup}$.
Thus it was concluded that an object of about Jupiter-mass was found.

That this object is also a planet stems from the difficulty of making a stellar companion (typical binary stars have about equal mass, but ratios of up to a factor of 10 are normal), and then evaporating it enough to loose not 100% of the mass, but just about 99%, which is another improbable mechanism.

Considering the last part of your question, stars don't just become 'small' at the end of their lives. Stars with significant mass-loss at the end of their lives loose this mass via strong stellar winds, which a) leave very visible planetary nebulae (which is absent in 51 Peg) and again b) leaving such a small blob of mass (compared to a hypothetical progenitor star) of a tiny fraction of the original mass is improbable, or even impossible (strong winds cease to exist at certain stellar masses/radii).

  • $\begingroup$ One doubt: How do we consider the above question for a more massive planet, like WASP-18b and the like? $\endgroup$
    – Maan
    Nov 16, 2019 at 13:57
  • 1
    $\begingroup$ @Maan: Well, if something like WASP-18b would have been the first HJ to be discovered, the scientific impact and the planetary nature might have been less clear. WASP-18b is a transiting planet (and was discovered as such), so it became clear that $\sin(i)=1$, and thus there was no doubt in the mass, which was only measured post-discovery. Another hypothetical $10 M_{\rm Jup}$ planet that is non-transiting would have had a realistic chance of being above the deuterium burning limit, as $sin(i)$ would be unknown then. $\endgroup$ Nov 16, 2019 at 14:35
  • 1
    $\begingroup$ @Maan: The issue with binary star formation is, that the less massive the companion is, the more one runs into a fine-tuning problem of explaining those low masses via cloud fragmentation. This was maybe still a discussion point at the time, but nowadays we know from statistics that a) around 1% of all stars have hot Jupiters (waaaay to much for fragmentation scenarios) and b) there is a 'brown dwarf desert', which is a gap in number populations as function of mass that clearly separates stars and planets, see also fig. 8 in this important article: arxiv.org/abs/astro-ph/0412356 $\endgroup$ Nov 16, 2019 at 14:42

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.