≡ Menu

55 Cancri A: Stable Orbital Solutions

We’re developing a model for the fascinating planetary system around the binary star 55 Cancri, a challenging task given the complexity of the inner system in particular. What we have here is a G-class star around which five planets are known to orbit and a distant M-dwarf at over 1000 AU. Have a look at the diagram below and you’ll see why the system, 39 light years away in the constellation Cancer, draws so much attention. It’s much more than the fact that direct measurements of the G-class star’s radius are possible at this distance, which have led to precise measurements of its mass, about the same as our Sun. It’s also the tightly packed configuration of the inner planets.


Image: An illustration of the orbital distances and relative sizes of the four innermost planets known to orbit the star 55 Cancri A (bottom) in comparison with planets in own inner Solar System (top). Both Jupiter and the Jupiter-mass planet 55 Cancri “d” are outside this picture, orbiting their host star with a distance of nearly 5 astronomical units (AU), where one AU is equal to the average distance between the Earth and the Sun. Credit: Center for Exoplanets and Habitable Worlds, Penn State University.

First discovered to be orbited by a giant planet in 1997, 55 Cancri A has been the subject of numerous studies in the years since. We have five planets in total, one a cold gas giant evidently similar to Jupiter and in a similar orbit and another, of particular interest, a ‘super-Earth’ in close proximity to the host star. This world, 55 Cancri e, was thought until 2011 to orbit the star in three days, but astronomers then determined that its complete orbit took less than 18 hours. The software developed by Penn State graduate student Benjamin Nelson and Eric Ford (Penn State Center for Astrostatistics) has pegged the mass of 55 Cancri e at eight Earth masses.

A quick note on nomenclature: The formal designation for the innermost world here should be 55 Cancri A e, with the other planets referred to accordingly. I’m following the just published paper on this work in referring to it as 55 Cancri e, without reference to the distant M-dwarf.

The transiting world is now known to have a radius twice that of Earth and a density about the same as our planet. Another glance at the diagram shows that this is a world far too hot for life as we know it, reaching temperatures in the range of 2300 Kelvin. The computations of Nelson and Ford draw the details of 55 Cancri e out of the motions of the giant planets 55 Cancri b and c, worlds that although orbiting outside the orbit of 55 Cancri e are still located closer to the star than Mercury is to our Sun. The new techniques help us understand how large planets like these can orbit so close to their star without collision or the expulsion of one of the two worlds.

The motion of the inner giant planets has to be accounted for to measure the detailed properties of the ‘super-Earth,’ and Ford notes that most previous work on this system had ignored their interactions. Nelson explains the significance of understanding the stability of their orbits:

“These two giant planets of 55 Cancri interact so strongly that we can detect changes in their orbits. These detections are exciting because they enable us to learn things about the orbits that are normally not observable. However, the rapid interactions between the planets also present a challenge since modeling the system requires time-consuming simulations for each model to determine the trajectories of the planets and therefore their likelihood of survival for billions of years without a catastrophic collision.”

1418 radial velocity observations from four observatories went into this work along with transit studies for 55 Cancri e, out of which orbital solutions stable for a minimum of 108 years emerge. The researchers evolved four- and five-planet models as they examined instabilities in the system, coupling their work with radial velocity observations to constrain the planet masses and orbital parameters that produce dynamically stable solutions. As the paper notes, “By combining a rigorous statistical analysis, dynamical model and improved observational constraints, we obtain the first set of five-planet models that are dynamically stable.” Another interesting finding: 55 Cancri d turns out to be “the closest Jupiter analog to date” in terms of orbital period and eccentricity.

The paper is Nelson et al., “The 55 Cancri Planetary System: Fully Self-Consistent N-body Constraints and a Dynamical Analysis,” Monthly Notices of the Royal Astronomical Society, published online 22 April 2014 (preprint). Also see this Penn State news release.


Comments on this entry are closed.

  • Lionel April 24, 2014, 10:46

    Reminded every week that it’s a great time to be an exoplanet researcher (or in my case just a casual hobbyist ;)

    BTW Paul I presume you know about this: http://www.bbc.com/future/story/20140423-return-of-the-nuclear-spaceship

    Kelvin Long interviewed by the BBC about Project Orion and Project Daedelus, with a link to Initiative for Interstellar Studies post on Centauri Dreams

  • Paul Gilster April 24, 2014, 12:58

    Yes, saw that story, Lionel, but thanks for checking. I get many good tips this way.

  • Michael April 24, 2014, 14:08

    A truly tortured world. Blasted by its unlovely Sun and bullied by its siblings tidal heating.


    It might have the last laugh though, it could be dragging them all in!

  • andy April 24, 2014, 18:08

    55 Cancri is not to be underestimated, I remember the very long integration times required when using the systemic console. It’s somewhat annoying to realise that the 1-day cutoff in the console hid the true period of planet e from the sight of everyone for so long – to be fair that kind of thing was fairly common before the robust detection of ultra-short period planets.

    Interesting to see that the eccentricity estimate for planet f is revised downwards, multi-planet systems tend to have lower eccentricities. As the paper notes it is located in the habitable zone but as a low-mass gas giant it does not seem particularly promising. Still no sign of additional planets between f and d despite this region being dynamically stable (perhaps including the outer part of the HZ).

  • Michael April 26, 2014, 3:19

    @andy April 24, 2014 at 18:08

    ‘Still no sign of additional planets between f and d despite this region being dynamically stable (perhaps including the outer part of the HZ).’

    This system is the end result of a migration process, it is most likely that they where further out and the orbit patterns where very different. It might be possible to back track the process but it looks even more complicated. That possible planet between f and d could have been flung out/absorbed or flung into the star. The star appears metal poor so it is unlikely in went into it and there is no dust disc that would indicate a collision, so it either didn’t exist or was flung out.

  • andy April 27, 2014, 7:54

    @Michael: I think that’s the first time I’ve seen 55 Cancri referred to as “metal-poor”. The star is substantially more metal-rich than the Sun, [Fe/H] around +0.3. The near-circular orbits of the detected planets don’t particularly suggest planet scattering either.

  • Michael April 27, 2014, 13:09


    Apologies Andy you are correct, I seen 0.31 (Fe/H) as the factor but I forgot it is a logarithmic representation, my mistake. It has a ‘metal’ enrichment of around twice that of the Sun which I found strange as it is quite old ~10 Gyr.
    Any ideas why or was it just the luck of the draw?

    ‘The near-circular orbits of the detected planets don’t particularly suggest planet scattering either.’

    you are correct again, the old data had f as quite eccentric

    new data

    Then there could be room for an as yet to be found small planet there, with any luck.

  • Ricardo Orsini (ROCA) May 6, 2014, 20:17

    One question:

    The arxiv article mentions 1418 high precision observations in the abstract.

    They do not mention “1418 hours of radial velocity observations”.

    Please check this.

  • Paul Gilster May 7, 2014, 10:29

    Ricardo, you are exactly right. I’ve changed the text to reflect what the arXiv paper says. Thanks for catching this!

  • ljk May 7, 2014, 12:01

    Habitable Planet Reality Check for 55 Cancri f moons

    by Andrew Lepage:


  • Aleksandar Volta July 30, 2014, 17:12

    I don’t see why ‘f’ couldn’t be a host of optimally large moon that has at least some kind of magnetic field and an atmosphere. Coupled with a hypothetical terran or superterran planet ‘g’ in orbit just after the gas giant, there could be a total of two HZ bodies in this system.