≡ Menu

New Discovery Around 55 Cancri

I want to get back to James Bickford’s antimatter study tomorrow, at which time I’ll set up the full report for download here. This work has already elicited plenty of response, both in comments and backchannel, so tomorrow we’ll talk about the mechanisms that create antimatter in our Solar System naturally (as opposed to what we do with particle accelerators), and also ponder how realistic missions to harvest such antimatter could be built around technologies currently in the pipeline.

Right now, though, the news conference on 55 Cancri is ongoing, the news from this system more and more interesting now that another planet, the fifth, has been discovered. This is the first system found with this many planets, although the assumption is there will be many, many others. But 55 Cancri, some 41 light years away in the constellation Cancer, bears some resemblances to our own Solar System. The farthest planet from the star is a gas giant about four times Jupiter’s mass orbiting every 14 years, which is not dissimilar to Jupiter’s 12-year orbit.

The new planet, discovered with data from Lick Observatory and the Keck Observatory in Hawaii, shows about 45 Earth masses (half the mass of Saturn) and orbits the star every 260 days. Interestingly enough, this fourth planet out from 55 Cancri lies in the system’s habitable zone, as shown in this slick animation. So here we go with the ‘habitable moons around giant planets’ discussion again, the assumption being that a large enough moon around this planet could retain liquid water and perhaps become a home to some form of life. The question is even more interesting given that the planets in this system, like those in ours, move in relatively circular orbits.

55 Cancri system

Image: The newest member of the 55 Cancri family – a massive planet, likely made of gas, water and rock, about 45 times the mass of Earth and orbiting the star every 260 days. This planet is the fourth out from the star, and lies in the system’s habitable zone (green). A habitable zone is the place around a star where liquid water would persist. Though the newest planet probably has a thick gaseous envelope, astronomers speculate that it could have one or more moons. In our own solar system, moons are common, so it seems likely that they also orbit planets in other solar systems. If such moons do exist, and if they are as large as Mars or Earth, astronomers speculate that they would retain atmospheres and surface liquid water that might make interesting environments for the development of life. Credit: NASA/JPL-Caltech.

Could 55 Cancri hold even more planets? Sure, and that might include small, rocky worlds too small for current detection. But even with the planets we now observe, the resemblances I spoke about above are noteworthy. No other system has a Jupiter-class planet in such a distant orbit, and in both systems, inner planets that are less massive than the outer ones are found. On the flip side, look at how close 55 Cancri’s three inner worlds are to their star. The closest, of Uranian mass, is on a three-day orbit. The second, not much smaller than Jupiter, orbits in about 15 days, while the third, a Saturn-class world, orbits every 44 days. That’s a lot of close-in activity compared to what we’re familiar with.

Even as the news conference proceeds, I’m thinking we’re living in the golden age of exoplanet research. I remember wondering once what it must have been like to do astronomy in the Edwin Hubble era, when the size of the universe itself was being so radically redefined and our understanding of the nature of distant objects completely altered. But this is better, with the prospect of moving from single planet discoveries to fully characterized systems ahead of us. Add the potential of space-based terrestrial planet finding missions and spectroscopic data on planetary atmospheres and I can only imagine the rush of new talent that’s going to be wanting to contribute to this field.

Related: Note this, from a recent report on the CoRoT mission at the 300-day mark (and sent along by our friend Vincenzo Liguori): “CoRoT is discovering exo-planets at a rate only set by the available resources to follow up the detections…” A golden age indeed.

Comments on this entry are closed.

  • tacitus November 6, 2007, 15:37

    Rumors are that the COROT team will be announcing their first batch of discoveries within weeks, if not days. This is exciting for (at least) two reasons:

    (a) since they are surveying a population of stars, if they can confirm a sizable number of discoveries in just a short time, we may be well on the way to getting a true understanding of just how many planets are out there waiting to be discovered.

    (b) preliminary tests have proved that their instruments are even more sensitive than originally spec’ed, and there is a good chance that they will be able to detect planets all the way down to near Earth-sized.

    I think Golden Age is an appropriate term, and the most exciting thing about it is that it has barely gotten underway. :D

  • Adam November 6, 2007, 16:07

    Hi Paul

    Jean Schneider’s Exoplanet Encyclopedia has the preprint available from the discovery team…


    …it seems unlikely that the new planet will have a habitable moon, but then the Earth/Moon binary is seemingly unlikely too. I think if the new planet has a habitable moon it will be a former Trojan planet captured by planet f.

  • andy November 6, 2007, 16:13

    Problem with habitable moons is the ~10^-4 mass scaling of gas giant satellite systems. If this ratio also applies to the 55 Cancri satellites then there might not be large enough moons around the habitable zone planet to support an Earthlike environment – if that kind of thing is even possible in the high radiation environment around a gas giant. Then again, the 55 Cancri system does contain a bunch of planets which are seriously overgrown compared to our own solar system, so who knows?

    Another thing to bear in mind is that 55 Cancri is a binary system, the secondary being a red dwarf. It would be very interesting to find out whether 55 Cancri B has a planetary system as well.

  • Enzo November 6, 2007, 16:21

    COROT preliminary report is here :

    It says, amongst the others that :
    “CoRoT is discovering exo-planets at a rate only set by the available
    resources to follow up the detections”

    To give an example of the follow up required, they give an example
    of raw data with a regular big spike and it says that it’s either a very
    large planet or a very small star. Presumably they need to characterize
    the star to understand its type and size and consequently deduce
    the size of the planet.


  • tacitus November 6, 2007, 17:01

    Thanks, Enzo, I hadn’t noticed the news item on their site. It’s a great teaser, and I’m certainly looking forward to the publication of their papers.

    It seems as though COROT is going to single handedly advance the science of both planets and stars to a remarkable degree.

    And this is just the beginning.

  • Administrator November 6, 2007, 17:13

    Well done, Enzo. That comment about being limited only by “…the available
    resources to follow up the detections” gives us an idea what we may expect from this splendid mission.

  • Zen Blade November 6, 2007, 17:30


    I was just looking at Wikipedia for Moon:Earth mass ratio, and then I looked at Saturn:Earth, and Mars:Earth.

    The Moon’s mass (relative to Earth) is about 1.2% , and Mars is about 10% of the Earth’s mass.

    So, if this new planet around 55 Cancri is about 40-50x the mass of Earth, there is a moon about 1% of the planet’s mass orbiting, that could mean a satellite roughly 40-50% of Earth’s mass.
    I’m sorry if everyone else already knew this, but I needed to look at the numbers first since often big announcements contain lots of optimistic speculation.

    The limitation (to my naive mind) being the fact that the largest moon we know of has a mass of only .025 Earths.

    Very cool news.
    -Zen Blade

  • andy November 6, 2007, 17:57

    Zen Blade: on the other hand, if you scale to the Titan:Saturn mass ratio or the Triton:Neptune one (gas giant satellite systems with most of the mass in a single large moon), you get an expected mass roughly equal to the Earth’s moon.

    Scaling to Ganymede:Jupiter or Titania:Uranus (gas giant satellite systems with most of the mass in multiple large moons) gives smaller expected mass, around 0.1-0.3 of a lunar mass.

  • Mark Atwood November 6, 2007, 19:15

    Forget “habitable moons”, from a planetology perspective, I want to know what a hot gas giant “looks” like!

    Most of the colorful look of Jupiter and Saturn are related to the fact that their visible light cloudtops are crypogenically cold. What would such a world look like when it’s Earth or Venus or Mercury temperature?!

    Do we even know that these heavy close-in planets are gas giants, and not huge rocky bodies? Maybe one of the oddities about the Sol system is that our inner rocky worlds are so small…

  • Adam November 7, 2007, 3:09

    Hi andy & Mark

    andy, your mentioning the observed mass limit of Jovian moons is why I said a captured Trojan planet. A Trojan planet in L4 or L5 is stable until its mass hits about 4.5% of the primary planet in that orbit – then it starts wandering along the orbit until the two interact.

    Mark, there have been quite a few papers on just what a “hot” Jovian will look like. The late John Whatamough references a few of them on his Extrasolar Visions web-site, so check them out. I suspect that such planets will be very blue, but there are all sorts of possible colours that might be dredged up by deep convection. The cloud patterns will be interesting – to say the least.

  • andy November 7, 2007, 9:36

    Adam: the issue is how to capture something that isn’t bound to the system – presumably you have to eject something comparable in mass to transfer momentum/energy to (it is interesting to note that Triton obeys the mass scaling despite being a captured object). I suppose a Trojan planet might just be going slow enough relative to 55 Cnc f for capture to work. It would also help if the hypothetical Trojan planet was itself a binary planet.

    So to get this somewhat unlikely scenario to work, we form a Trojan planet, whack it with something big to form a large moon, which might also help in knocking it out of the stable region. The Trojan planet and its moon then migrates towards the gas giant, and ends up captured in orbit. Hmmm…

    Whether the Trojan points would be stable enough to form Earth size planets with all the other massive planets in the system migrating around is another question, particularly if the planets went through resonances with each other.

  • ljk November 7, 2007, 12:56

    Extreme Habitability: Formation of Habitable Planets in Systems with Close-in Giant Planets and/or Stellar Companions

    Authors: Nader Haghighipour

    (Submitted on 6 Nov 2007)

    Abstract: With more than 260 extrasolar planetary systems discovered to-date, the search for habitable planets has found new grounds. Unlike our solar system, the stars of many of these planets are hosts to eccentric or close-in giant bodies. Several of these stars are also members of moderately close ($ less than $40 AU) binary or multi-star systems. The formation of terrestrial objects in these “extreme” environments is strongly affected by the dynamics of their giant planets and/or their stellar companions. These objects have profound effects on the chemical structure of the disk of planetesimals and the radial mixing of these bodies in the terrestrial regions of their host stars. For many years, it was believed that such effects would be so destructive that binary stars and also systems with close-in giant planets would not be able to form and harbor habitable bodies. Recent simulations have, however, proven otherwise. I will review the results of the simulations of the formation and long-term stability of Earth-like objects in the habitable zones of such “extreme” planetary systems, and discuss the possibility of the formation of terrestrial planets, with significant amounts of water, in systems with hot Jupiters, and also around the primaries of moderately eccentric close binary stars.

    Comments: 6 pages, 5 figures, to appear in the proceedings of “Bioastronomy 2007”, ed. Meech et al

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0711.0782v1 [astro-ph]

    Submission history

    From: Nader Haghighipour [view email]

    [v1] Tue, 6 Nov 2007 00:31:59 GMT (667kb)


  • David November 8, 2007, 9:42


    Also check out Greg Laughlin’s site at http://www.oklo.org where he also discusses the colours of hot jupiters. It turns out that at least some of them will have very low albedos and may look purplish, so think dark purple.

  • andy November 8, 2007, 11:22

    One thing I’m wondering about is just how planets c and f failed to accrete enough hydrogen/helium to reach Jupiter masses, when planets b and d (interior and exterior to the two sub-Saturns) did.

    Perhaps these two planets formed later, when the protoplanetary disc contained less gas, in which case they might be super-Neptunes rather than mini-Saturns.

  • Ronald November 8, 2007, 21:33

    @andy: c and f also have significantly greater eccentricity. Is there a relation? I mean, do they belong to two different ‘populations’? One group which originated in situ and one which migrated? or, as andy suggests, groups which formed at different stages.

    Makes one wonder what other (smaller) planets may still be hiding in between them. If two such large planets (b and c) can be only 0.12 AU apart (c and f 0.5 AU apart), there must be place for a whole chain more. Maybe this will appear to be typical of high metallicity stars (55 Cnc has about twice solar metal.).

  • Dave Moore November 9, 2007, 2:58

    The spacing of the planets in the 55 Cancri system seems to be reasonably regular; although, there is room for a planet between F and D. If you take the ratio of the orbital distance of a planet to the distance of the next one in you get figures of 3.0, 2.1, 3.25 & 7.2. The square root of 7.2 is 2.7 a figure in line with the other spacings. This would put the missing planet at about 2.1 au.
    It is interesting to note the distance ratios are all greater than 2 and more often 3. This compares with our solar system where each planet is always less than twice the distance of the one interior to it, the closest distance ratio being between Earth and Venus with a ratio of 1.4 and the greatest being between Uranus and Saturn, nearly 2. I would guess that the greater distance ratio of the 55 Cancri system allowed the system to maintain so many large planets in stable orbits.
    I took a look at the other multiple planetary systems. Most of them had at least some planets in wildly scrambled orbits of high eccentricity, making this type of analysis impossible.
    The red dwarf Gliese 581 has distance ratios of 1.8 if you assume there is a planet between c & d. Of course, both 581c and d have moderately eccentric orbits so this may be an indication that the planet inbetween was ejected. The only other system that I could see any regularity in was the K dwarf HD 69830 which has 3 planets of 11, 12 & 18 earth masses in tight, fairly circular orbits. Their distance ratios are 2.4 and 3.3. It may be that each planetary system may have is own spacing ratio. Obviously a system with tight spacing ratios isn’t going to maintain too many Jovian sized planets in stable, circular orbits. All this is conjecture of course as we don’t have a large enough sample.
    Getting to the subject of orbital eccentricity. I regard any orbit with an eccentricity of under 0.1 as being basically circular. Mars has an orbital eccentricity of 0.09. So to my way of thinking only 55 Cancri f would be in a different orbital class, and even then its orbit is only as eccentric as Mercury’s.


  • andy November 9, 2007, 9:52

    @Dave Moore: yes, this kind of system does allow for checking whether relationships similar to the Titius-Bode “law” hold in other solar systems. I think I managed to find a fairly decent logarithmic relationship between the Mu Arae planets, but you have to assume there are several objects/asteroid belts in the gaps between the first and second, and third and fourth planets.

  • ljk November 30, 2007, 11:10

    Retired A Stars and Their Companions II: Jovian planets orbiting kappa Coronae Borealis and HD167042

    Authors: John A. Johnson, Geoffrey W. Marcy, Debra A. Fischer, Jason T. Wright, Sabine Reffert, Julia M. Kregenow, Peter K. G. Williams, Kathryn M. G. Peek

    (Submitted on 28 Nov 2007)

    Abstract: We report precise Doppler measurements of two evolved stars, kappa CrB (HD142091) and HD 167042, obtained at Lick Observatory as part of our search for planets orbiting intermediate-mass subgiants. Periodic variations in the radial velocities of both stars reveal the presence of substellar orbital companions. These two stars are notably massive with stellar masses of 1.80 Msun and 1.64 Msun, indicating that they are former A-type dwarfs that have evolved off of the main sequence and are now K-type subgiants. The planet orbiting kappa CrB has a minimum mass Msini = 1.8 Mjup, eccentricity e = 0.146 and a 1208 day period, corresponding to a semimajor axis of 2.7 AU. The planet around HD167042 has a minimum mass Msini = 1.7 Mjup and a 412.6 day orbit, corresponding to a semimajor axis of 1.3 AU. The eccentricity of HD167042b is consistent with circular (e = 0.027+/-0.04), adding to the rare class of known exoplanets in long-period, circular orbits similar to the Solar System gas giants. Like all of the planets previously discovered around evolved A stars, kappa CrBb and HD167042b orbit beyond 0.8 AU.

    Comments: 8 pages, 3 figures, 4 tables, ApJ Accepted

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0711.4367v1 [astro-ph]

    Submission history

    From: John Johnson [view email]

    [v1] Wed, 28 Nov 2007 05:59:30 GMT (43kb)


  • ljk December 20, 2007, 0:46

    Formation and transformation of the 3:1 mean-motion resonance in 55 Cancri System

    Authors: Li-Yong Zhou, Sylvio Ferraz-Mello, Yi-Sui Sun

    (Submitted on 19 Dec 2007)

    Abstract: We report in this paper the numerical simulations of the capture into the 3:1 mean-motion resonance between the planet b and c in the 55 Cancri system. The results show that this resonance can be obtained by a differential planetary migration. The moderate initial eccentricities, relatively slower migration and suitable eccentricity damping rate increase significantly the probability of being trapped in this resonance. Otherwise, the system crosses the 3:1 commensurability avoiding resonance capture, to be eventually captured into a 2:1 resonance or some other higher-order resonances. After the resonance capture, the system could jump from one orbital configuration to another one if the migration continues, making a large region of the configuration space accessible for a resonance system. These investigations help us understand the diversity of resonance configurations and put some constrains on the early dynamical evolution of orbits in the extra-solar planetary systems.

    Comments: 6 pages with 2 figures. Submitted for publication in the proceedings of IAU Symposium No.249. A paper telling much more details than this paper is under preparing

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0712.3138v1 [astro-ph]

    Submission history

    From: Liyong Zhou [view email]

    [v1] Wed, 19 Dec 2007 10:03:47 GMT (107kb)


  • ljk December 31, 2007, 3:16

    Five Planets Orbiting 55 Cancri

    Authors: Debra A. Fischer, Geoffrey W. Marcy, R. Paul Butler, Steven S. Vogt, Greg Laughlin, Gregory W. Henry, David Abouav, Kathryn M. G. Peek, Jason T. Wright, John A. Johnson, Chris McCarthy, Howard Isaacson

    (Submitted on 23 Dec 2007 (v1), last revised 27 Dec 2007 (this version, v2))

    Abstract: We report 18 years of Doppler shift measurements of a nearby star, 55 Cancri, that exhibit strong evidence for five orbiting planets. The four previously reported planets are strongly confirmed here. A fifth planet is presented, with an apparent orbital period of 260 days, placing it 0.78 AU from the star in the large empty zone between two other planets. The velocity wobble amplitude of 4.9 \ms implies a minimum planet mass \msini = 45.7 \mearthe. The orbital eccentricity is consistent with a circular orbit, but modest eccentricity solutions give similar \chisq fits. All five planets reside in low eccentricity orbits, four having eccentricities under 0.1. The outermost planet orbits 5.8 AU from the star and has a minimum mass, \msini = 3.8 \mjupe, making it more massive than the inner four planets combined. Its orbital distance is the largest for an exoplanet with a well defined orbit. The innermost planet has a semi-major axis of only 0.038 AU and has a minimum mass, \msinie, of only 10.8 \mearthe, one of the lowest mass exoplanets known. The five known planets within 6 AU define a {\em minimum mass protoplanetary nebula} to compare with the classical minimum mass solar nebula. Numerical N-body simulations show this system of five planets to be dynamically stable and show that the planets with periods of 14.65 and 44.3 d are not in a mean-motion resonance. Millimagnitude photometry during 11 years reveals no brightness variations at any of the radial velocity periods, providing support for their interpretation as planetary.

    Comments: accepted to ApJ

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0712.3917v2 [astro-ph]

    Submission history

    From: Debra A. Fischer [view email]

    [v1] Sun, 23 Dec 2007 14:22:09 GMT (350kb)

    [v2] Thu, 27 Dec 2007 09:48:27 GMT (350kb)


  • ljk March 18, 2008, 0:06

    The Exo-planetary System of 55 Cancri and the Titius-Bode Law

    Authors: Arcadio Poveda, Patricia Lara

    (Submitted on 14 Mar 2008)

    Abstract: The recent discovery of a fifth planet bound to 55 Cancri (Fischer et. al 2007) motivated us to investigate if this exo-planetary system fits some form of the Titius-Bode (TB) law. We found that a simple exponential TB relation reproduces very well the five observed major semi-axis, provided we assign the orbital n = 6 to the largest a. This way of counting leaves empty the position n = 5, a situation curiously reminiscent of TB law in our planetary system, before the discovery of Ceres.

    The application of an exponential TB relation to 55 Cancri allows us to predict the existence of a planet at a = 2.0 AU with a period of P = 1130 days located within the large gap between a = 0.781 AU (P = 260 days) and a = 5.77 AU (P = 5218 days). With less certainty, we also predict a seventh planet at a = 15 AU with P = 62 years.

    Comments: Accepted for publication in RevMexAA

    Subjects: Astrophysics (astro-ph)

    Cite as: arXiv:0803.2240v1 [astro-ph]

    Submission history

    From: Arcadio Poveda [view email]

    [v1] Fri, 14 Mar 2008 20:12:48 GMT (113kb)


  • Warren Platts April 11, 2009, 22:06

    I discovered and published this pattern months before Poveda and Lara on November 13, 2007:

    “However, five exoplanets have been discovered around 55 Cancri A. Amazingly, the semimajor axes are well predicted by the following simple relation:

    a = 0.039e^{n-1}

    where a is the semimajor axis, e is the natural logarithm constant (2.7 …), n is the number of the planet starting from it’s sun, and the 0.039 is the semimajor axis of the closest planet in AU’s.

    . . .

    55 Cancri

    1. 0.038 — 0.039 — 0.0000010
    2. 0.115 — 0.106 — 0.0000808
    3. 0.240 — 0.288 — 0.0023207
    4. 0.781 — 0.783 — 0.0000055
    5. ????? — 2.129 — ??????????
    6. 5.770 — 5.788 — 0.0003281
    7. ????? — 15.73 — ??????????

    . . .

    It is striking that the very first extrasolar test of Bode’s Law-like relations agrees with the predictions better than our own solar system does.

    If Bode’s Law like spacings are ubiquitous, that suggests that logarithmic spacings are not in fact merely coincidental, but result from a deep, underlying physical explanation.