Spitzer sees the glow of a boiling planet

By Phil Plait | May 8, 2012 2:04 pm

Since the first planet was discovered orbiting another Sun-like star in 1995, nearly 800 more have been discovered. Only a handful have been directly detected: most are discovered by their influence on their star, either by tugging it or blocking its light as the planet orbits (at the bottom of this post is a gallery of images of exoplanets detected in these ways). But some have been directly seen: either glowing by their own light, reflecting that of their star, or — ironically — seen when they’re not seen.

Say what? OK, this takes a sec to explain, but it’s cool.

The star 55 Cancri hosts at least 5 planets. Located 40 light years away, it’s one of the closer planetary systems, and has been intensely studied. One of the planets, 55 Cancri e, is bizarre: it’s twice the diameter of the Earth and has 8 times our mass. It’s thought to have a dense core surrounded by water… but Earth-like it ain’t. It orbits its star in a very tight orbit, circling it once every 18 hours. It’s so close to the star that the surface temperature is probably around 1700°C — or 3100°F! That’s hot enough to melt lead.

So yikes. If it does have water, it’s in the form of a weird super-heated steam only held to the planet due to its strong gravity. Even then, the atmosphere may be boiling away like a gigantic comet. So again, this isn’t like Earth at all. Even Venus isn’t this unpleasant, and on Venus it rains sulfuric acid.

Anyway, an object at that temperature will glow in the infrared, quite strongly. If it were sitting all by itself in space, it would be easy to see. However, it’s sitting next to a star which is millions of times brighter, making it a significantly more difficult target.

… but not impossible. The orbit of the planet 55 Cancri e is edge-on as seen from Earth, so every 18 hours the planet crosses in front of the star. When it does, the starlight dips a bit because the planet blocks it. However, if you wait 9 hours, the planet will pass behind the star. Even though the star is far brighter, that means there’s still a small dip in the total infrared light emitted by the system (the folks at Spitzer have a nice graphic showing how this works).

Using the infrared-viewing Spitzer Space Telescope, astronomers observed the star 55 Cancri and waited for the planet to go behind the star. What they saw was this:

The points represent the amount of infrared light they measured (at 4.5 μ, well outside what our eyes can see), and the x-axis is time. As you can see, the light from the star dipped a bit. Not very much! The depth of the eclipse is only about 0.01%, but that’s enough to see. And by observing the event over and over again they were able to get pretty good statistics, and confirm they are seeing the light from the planet being blocked by the star. And while this has been done for other exoplanets before, this is the first time this technique has been successful for a "super-Earth" like 55 Cancri e.

How awesome is that? Even though the planet is only emitting light very feebly — it’s 400 trillion kilometers (240 trillion miles) away, after all — the dull glow of its superheated surface can still be detected if you look at the right time, and you’re ready to measure when it goes away.

And that’s what I meant by seeing a planet by not seeing it. It’s just like the saying goes: you don’t know what you have until it’s gone.

Related Posts:

Nearby "earth-like" planet: not so much
Astronomers find 5 planet system!
Dense exoplanet gets the lead out and in

[Below is a gallery of exoplanets that have been directly imaged using telescopes on ground and in space. Click the thumbnail picture to get a bigger picture and more information, and scroll through the gallery using the left and right arrows.]



Comments (29)

  1. Valdis Kletnieks

    http://www.nasa.gov/mission_pages/spitzer/news/spitzer20120508.html is NASA’s take on it, but it leaves me confoozled…

    On the one hand, it says:

    “The results reveal the planet is likely dark, and its sun-facing side is more than 2,000 Kelvin (3,140 degrees Fahrenheit), hot enough to melt metal.

    The new information is consistent with a prior theory that 55 Cancri e is a water world: a rocky core surrounded by a layer of water in a “supercritical” state where it is both liquid and gas, and topped by a blanket of steam.”

    but the very next paragraph says:

    “indicating the planet probably does not have a substantial atmosphere to carry the sun’s heat to the unlit side.”

    If there’s not much atmosphere, what’s providing the pressure to keep water in a supercritical state at 2000C? That takes a *lot* of PSI.

  2. Spocko

    @Valdis – Gravity?

  3. chief

    Forgetting the effects of the solar wind on a planet this close in, I wonder how much of the atmosphere is being dragged off the planet by its rapid orbit around the star. At what speed would a “earth” have to travel to leave the atmosphere behind.

  4. Ryan

    Small point but 2700c is way, way higher than the melting point of lead. It’s higher than the melting point of most metals. In fact, it’s higher than the boiling point of many metals, although likely not at the pressures you see on this planet.

  5. Matt

    If you want to help NASA analyze light curves in their search for extrasolar planets, visit planethunters.org and become a Planet Hunter.

  6. amphiox

    Well, a “blanket of steam” should count as an atmosphere!

    But “substantial” is a relative word. Not thick enough to efficiently carry heat around does not mean zero. Even the moon doesn’t have zero atmosphere.

  7. Chris

    @1 Valdis

    To have water in the supercritical state you need pressures above 217 atmospheres and temps above 617 K (650 F). So I would count that as a substantial atmosphere. As to the part which “confoozled” you. Reading the rest of the paragraph

    55 Cancri e the closest to the star and tidally locked, so one side always faces the star. Spitzer discovered the sun-facing side is extremely hot, indicating the planet probably does not have a substantial atmosphere to carry the sun’s heat to the unlit side.

    My guess here is that since the planet is tidally locked, one half of the planet is getting sand blasted by the star and any atmosphere has been pushed to the dark side of the planet. That sounds weird and I’m probably wrong, but that’s what I got from that paragraph.

  8. eyesoars

    Iron melts at only 1535 °C. Even some of the refractories are above their (1 atm) melting points: Tungsten: 3422 °C (the highest MP), Rhenium: 3186 °C, and Tantalum: 3017 °C, are still solid, but Molybdenum: 2623 °C, Niobium: 2477 °C would be liquid. Osmium (3045 °C) would be solid, but iridium (2410 °C) liquid.

    Lead *boils* at a mere 1750 °C (@ 1 atm).

  9. Chris

    @7 eyesores

    As for the metals you list, that’s just for the pure metal. On a planet they’d be mixed with other elements which would lower the melting point.

  10. arabwhipmonk

    Ended up signing up for Zooniverse after reading the post and comments. Thanks @Matt and @BadAstronomer

  11. scotty

    ok. the author made a little mistake. the nasa webpage says 2000K.
    2000 kelvin is about 1700 celsius so 2700 celsius was a mistype and then the mistype was correctly converted to fahrenheit

  12. For those asking, the temperatures I quoted are from an earlier post I wrote about this planet. In the case of that post, and the newer numbers the press release had, both are calculated making various assumptions, and are estimates. I don’t know which is likely to be more accurate, so for now I’ll leave things as is until I now more.

  13. flip

    What I love about this is that it’s a (relatively) simple solution to a complex problem. How to detect planets so far away? Just look for a dip in the star’s light. That’s some serious simplicity to a crazy complex problem.

  14. Artor

    Flip @13
    The problem is that it only works if the planet’s orbit lines up with ours. Otherwise, there is never a transit for us to observe. It’s kinda amazing that it has worked as often as it has, although I suppose most systems ecliptical planes would align approximately with the galaxy’s. Am I wrong on that Phil? Do we have any idea?

  15. Robert

    #Flip: Well, ours doesn’t! If you look up, you can seethe plane of the milky way, arcing across the sky fairly close to north-south. Our planet’s orbit is [up to, depending on the time of year,] 27° from east-west. So the orbital plane of our solar system is tilted far from the plane of the Milky Way.
    If it was lined up, the milky way would lie along the ‘ecliptic’, which is the line that the other planets (approximately) follow. So when you see Jupiter and Saturn shining brightly, far from the Milky Way in the sky, you again see how far our solar system is tilted with respect to our galaxy.

  16. goob

    It orbits a star in 18 hours?!! How fast is that thing travelling?

  17. flip

    @14 Artor & @15 Robert

    This is true… Taken from the perspective of someone who works in the arts though, and has little education in science other than high school classes… it’s still pretty darn awesome considering the difficulty of finding planets in other galaxies. Artor you’re right: I’m amazed watching the light dip has worked as often as it has. I guess that’s the advantage of having a shirtload of galaxies and stars to look at. Eventually you’ll see something given time and patience.

  18. Why do people always use lead as a ‘wow that’s hot!’ metal? The surface will melt iron and copper, it would literally glow with heat, I believe lead would be near boiling point there. (1740C)

  19. amphiox

    Artor, I don’t think there is any reason to think that planetary systems will line up with the galactic ecliptic. Our own does not.

    In fact I think that the distribution of alignments is basically random.

  20. Messier Tidy Upper

    Superluminous – beyond merely brilliant – work by Spitzer. :-)

    What a remarkable and unfathomably alien planet.

    Twice Earth’s diameter, eight times earth’s mass, a year that’s less than one of our days,

    Composed mostly, probably of superheated water under pressures beyond imagining.

    Perhaps with a layer of diamond somewhere deep within from the crushed to the max carbon.

    Could 55 Cancris e have a plasma core or metallic hydrogen layer like that within Jove?

    Could it have (like Earth to some degree and Jupiter to much more of a degree) a strong source of internal heat burning it from within as well as without?

    Could its dark side actually be frozen at the anti-stellar point whilst the hot side atmosphere simultanously bubbles and then streams away into space comet-like? A hidden zone of ice wrapped inside superheated fire and ethereal gas.

    What storms and bands and climatic zones might 55 Cancris e have?

    So much strangeness and wonder and so marvellous that we can find even so much as we know about it given its ungraspably vast distance away, comparitively microscopically small size and star-kissing location.

    Is 55 Cancris e really best called a “Super-Earth” or a “Gas dwarf” or perhaps another term is needed – Fire dwarf or Molten dwarf perhaps? 😉

    Whoah! What. A. Planet! 8)

  21. chris

    So we’ve just discovered hell? 😉

  22. amphiox

    Could its dark side actually be frozen at the anti-stellar point whilst the hot side atmosphere simultanously bubbles and then streams away into space comet-like? A hidden zone of ice wrapped inside superheated fire and ethereal gas.

    A frozen area on a water world means a transition somewhere between ice and water.

    And that means a potentially habitable region, even if just a sliver in width!

    On a world like THAT.

    Wouldn’t that be mind boggling?

  23. amphiox

    Is 55 Cancris e really best called a “Super-Earth” or a “Gas dwarf” or perhaps another term is needed – Fire dwarf or Molten dwarf perhaps?

    Autoclave dwarf?

  24. Georg

    There is some very layish thinking here about gas phase
    physics. An atmosphere will not enhance transport of heat
    but lower it, as far as water vaporisation is concerned.
    Without some “permanent gas” water vapor would stream
    from our tropic seas to the poles and freeze out there.
    Think of the polar caps of Mars!
    The only difference is, that due to the very low amount of
    carbon dioxide and low overall pressure it sublimes from
    pole to pole directly.

  25. @ amphiox (#23) that is a scenario that has been contemplated numerous times. I think that the hypothetical transition point on a planet like 55 Cancri e would be well around towards the dark side of the planet. Even so, the situation would be very difficult to model given how bloody close that thing is to its parent star.

    And for people wondering about transitions, when StackExchange had an astronomy section, I covered this in a question. All the questions got moved over to Physics, but it’s still an interesting question about transistions. Granted, the percentage is much higher for closer in planets (due to geometry), but still:


  26. Messier Tidy Upper

    For those who might want to go outside, look up and see the planetary system – or at least its primary* sun Rho-1 cancris aka 55 Cancris itself check out the photographic finderchart here :


    Via Kaler’s Stars website which also has more information on that star and its planets including that 55 Cancris e is perhaps as dense as iron overall! :-)

    There’s a wikipage here :


    too with more. :-)

    Slightly off-topic here – & hope its okay to mention – this :


    Interesting blog post via Stephen “DarkSyde” Andrew ‘s Zingularity blog on exoplanetary implications of Hot Jupiters and hot & cold earths. :-)


    * * 55 Cancris does actually have a stellar companion too – an M6 red dwarf star very distant from it but still gravitationally bound.

  27. Messier Tidy Upper

    @24. amphiox :

    “Is 55 Cancris e really best called a “Super-Earth” or a “Gas dwarf” or perhaps another term is needed – Fire dwarf or Molten dwarf perhaps?”
    Autoclave dwarf?

    Yep, that’d be apt. :-)

    @23. amphiox :

    A frozen area on a water world means a transition somewhere between ice and water. And that means a potentially habitable region, even if just a sliver in width!
    On a world like THAT. Wouldn’t that be mind boggling?

    Very! 8)

    Talk about having a restricted habitat though!

    @26. Larian LeQuella : Cheers. :-)


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar