Anybody wanna peanut?

By Phil Plait | April 16, 2007 7:42 pm

I love it when scientists study an object I can truly sink my teeth into.

The image above is of RS Oph, a weird star some 5000 light years away. The star itself is in the center of those odd rings, and in fact it isn’t an it, it’s a they. It’s actually a binary star, two stars orbiting each other. But neither star is normal. One, the primary (because it’s brighter), is a red giant: an old star, maybe like the Sun will be in a few billion years. It’s run out of hydrogen to fuse in its core, and as the core began to shrink it got hotter. The extra heat was dumped into the star’s outer layers… and they responded as any gas does to extra heat: they expanded. So the star has puffed up. But all that extra surface area of the enlarged star means the heat is spread out over more real estate, so overall the star cools off. It turns red. Hence the name red giant.

See? Weird.

Now, the companion has roughly the same mass, but is far, far smaller. It also used to be a star like the Sun, though more massive. It too went through a red giant phase, but it lost its outer layers over thousands of years, ejecting them in a super-solar wind. All that’s left now is the hot compressed core of the star, roughly the size of the Earth but with the mass of a full star. It’s very hot, glowing white, but very small, so it’s called a white dwarf.

Also weird.

But hang on! The white dwarf is so close to the red giant that it is actually drawing matter off the outside of the giant, gravitationally sucking it down like it has one of those curly crazy straws. The matter piles up on the white dwarf, and gets really squeezed due to the tremendous gravity of the dense star. If enough matter piles up, it can actually undergo catastrophic fusion, like a thermonuclear bomb. BANG! Matter is ejected at millions of kilometers per hour, and the system suddenly gets very, very bright.


Actually, not really. This event, called a dwarf nova, happens all the time in the Galaxy. There are lots of binaries similar to RS Oph, and they undergo these nova events too. The last one for RS Oph happened in February 2006, and this dim star suddenly got bright enough to see with the unaided eye, which is pretty cool (it rained solidly here last year and I never got a chance to see it before it dimmed).

However, RS Oph is not exactly like other binaries. The red giant is also blowing out a big, dense wind, and the white dwarf is sitting inside of it, essentially in the atmosphere of the red giant. When the dwarf nova occurred last year, the exploded material screamed out and slammed into the red giant wind, heating it to hundreds of millions of degrees. Within days, astronomers training their ‘scopes on the system were greeted by a cigar-shaped expanding cloud of gas.

Weird? Yeah, this time it is weird.

Right away, that cigar-shaped nebula was a clue that something truly interesting was going on. If the explosion had happened in deep space, it would have been spherical, expanding equally in all directions. The cigar-shape meant that something else was happening.

That something else was the red giant wind. Normally, a wind will expand in all directions, forming a sphere around the star (the Sun’s solar wind does this, more or less). But the white dwarf and the red giant orbit around each other, so the red giant makes a big circle as it does so. The wind coming from it will prefer to blow out in the plane of that orbit. Think of it this way: if you take a bucket of water and spin it around you, the water that spills out will tend to fly away from you in the plane of the circle the bucket makes, right? Same thing here.

So the wind is denser in the plane than toward the poles of the orbit. And it’s in this wind that the white dwarf sits. So when it went bang, the matter flying outwards faced a tougher haul in the plane of the orbit, where the matter is denser, than along the poles, where the matter is less dense. The exploded material moved faster up and down, if you will, so it elongated into a cigar shape!

Weirder and weirder. But now comes the weirdest part. 155 days after the explosion, a team of UK astronomers pointed Hubble at the system, and took that funny image at the top of this post. What’s all that about?

As the matter from the explosion continued to expand, it pushed against that old red giant wind. It also expanded sideways, since it’s hot, inflating like a balloon. Eventually, after a few months, it blew two huge bubbles into the relic wind, two lobes of matter. The lobes are really like shells, or soap bubbles: thin shells of matter compressed by the expanding hot gas inside.

We can model very easily what something like this would look like from the side:

Mmmmm, peanut. Actually, it’s a nephroid, a type of epicycloid. And it looks a lot like a soap bubble!

Now, we also know that the lobes of RS Oph are tilted to us by about 40 degrees, so we can model that too:

Hmmm, looking familiar. But the actually Hubble image of RS Oph can just barely resolve the nebula, so let’s fuzz out the model to represent how well Hubble would see it:

Aha! Let’s see them side-by-side, model on the left and the Hubble image on the right:

Cool, huh? It’s a match! Looking at that, it’s safe to say we have a pretty good grasp of what’s going on here.

Let me repeat that: we get this object. It’s 5000 light years away, it’s a binary, it’s two old stars, one puffy and bloated, the other tiny and compressed, and one is eating the other while the other blows off a wind, and the tiny one periodically explodes, and forms a peanut-shaped soap bubbly nephroidal shell, which we can observe and model simply using undergrad-level math. And with all that, we can see we’re pretty close to spot on with the models!

Ah, science. It works.

Now, as a postscript, I’m kicking myself. When I was in grad school I observed Supernova 1987A with Hubble, and it too had this double-lobe thingy going on. I spent months modeling it using the exact same technique these astronomers did on RS Oph. I knew all about those peanut-shaped nephroids (which is why I was able to tell you about them), as well as shells and trying to match up the models and images. In other words, had I stayed on top of this event with RS Oph, I could have predicted the image would look like this.


Bonus points to the first person who figures out the title to this post.

Comments (30)

  1. Gene

    The title is from the Princess Bride. “no more rhymes, now, I mean it!” “Anybody want a peanut?”

  2. ABR

    No more rhymes now…I mean it!

  3. ABR

    Rats! How did you do that, Gene? My browser refresh strikes again — you win.

  4. George

    Cool story and nice explanation. I am a little surprised that 1987A, a type II sn, would have lobe similarities.

  5. How far apart ARE binary stars, on average? I know the differing masses must make for big differences, but there must be some kind of average…



  6. LurchGS


    well, there is (not that I know it). If you have any whole number of values, there’s an average. The question is, I think, other than just cool factor, is there anything that the average would tell us? Being such a professional astronomer that I have to look up how to spell it, my bet is on ‘no’. I’d be more interested in the ends of the curve (and it appears that this pair is pretty close to the short end)

    Never get involved in a land war in Asia and never get between the BA and his stars..

    wait.. did I get that right?

  7. Supernova


  8. How many AU apart are the stars? How would they compare with the size of our solar system? And could any existing exoplanets in the system effect the structure of the gas shells and actually be discovered from their gravitationally & / or magnetically pertubing effect on the structure?

    BTW. If I recall right the red supergiant Betelgeux (Alpha Orionis – the tenth brightest star although sometimes brighter than the ninth as its a slight variable) actually has a star orbiting _INSIDE_ its diffuse outer layers.

    I also vaguely recall reading somewhere that Sanduleak (catalogue numbers) the blue supergiant star that went supernova in 1987 actually merged with a close compnanion star during its red supergiant phase, the merger turning it into the blue variant of supergiant inwhich state it finally blew up. Any chance of a merger -&perhaops subsequent supernovae in RS Ophiuchi does anyone know?

    Thanks BA -excellent news & very well elucidated! (Explained.)

  9. HawaiiArmo

    I got the Title. It’s a Type 1A supernova. That’s the type used by astronomers to determine the distance of distant galaxies as well as the rate of expansion (and proof of the accelaration in the last 6 billion years of the universe).

  10. John Oliver

    Binary star orbital periods have a wide range of values. Observed periods are as short as 5 minutes (the stars are separated by about 1/5 the distance to the Moon) and as long as 300+ years.

  11. DennyMo

    So if it’s called pareidolia when people see faces on their grilled cheese sandwiches, what’s it called when people extrapolate wildeyed stories about stellar constitution and behavior from what looks like a 20 pixel by 20 pixel fuzzy picture of… something?

  12. Liam

    I just wanna point out that at the top of your page Google has kindy provided a link to a site selling flights to Mars. Pretty cool.

  13. Why weird? It’s normal, things happen in cosmos.

  14. Oh yea, first read, then comment…
    You (BA) have a nice way of talking about these phenomena.

  15. You keep using that word. I do not think it means what you think it means.

  16. Irishman

    DennyMo, what makes you think that everything described about RS Ophiuchi comes from that one picture? Don’t you think that astronomers have been studying it for some time? Maybe looking at it before the recent explosion?

  17. PK

    Bah Rose, I wanted to post that comment…

    Very cool blog post, Phil. Did you do the modeling yourself?

  18. Cindy

    Yeah, a nice description of a cataclysmic variable. Thanks for blogging about it, Phil!

    In answer to previous posts, cataclysmic variable periods tend to vary between 1 to 2 hours and then are usually greater than 3 hours. The longer the period, the further apart the stars are.

    I don’t know off the top of my head the orbital period of RS Oph, but the separation tends to be less than 1 AU for the shorter period cataclysmic variables. I don’t have time to calculate it right now but could probably do it as I did my PhD thesis on cataclysmic variables, though not on RS Oph.

  19. That Neil Guy

    This post gave me a warm fuzzy feeling inside. When I was in high school, I wrote a letter to Isaac Asimov after reading one of his essays about recurring nova, as he called them. I wondered what would happen to the other star in the system, how it would be affected. And Dr. A kindly sent me a postcard saying that the star would undego some surface heating but would otherwise come out of it okay. Anyway, this post made me think about that and how much I enjoyed reading Dr. A. I’m not going to say that BA fills the void left by Dr. A’s death, but it sure helps that someone is out there still trying to explain to us the wonders of the cosmos.

  20. peter

    >>Rose Explain please if that was to me. :)

  21. Donnie B.

    BA, you’d make a wonderful Dread Pirate Roberts.

    On the issue of an “average” binary distance, I don’t think there’s a Gaussian distribution for such things, and therefore no “most likely” distance. Some binaries are very close (I think most recurring nova systems would be this kind) while others are much farther apart than, say, Neptune and the Sun.

    This leads to several possible configurations for planets in such systems. If the binary pair is close together, there could be planets orbiting both stars. If they’re far apart, each star could have its own planetary system… and there could still be an even more distant “Oort cloud” orbiting both stars.

  22. J Jonah Jansen

    Peter, Rose was referencing more Princess Bride dialog re; Supernova’s comment – “Inconceivable”. A must see… if you ask me. ;-)

  23. Blaine

    I do not think that word means what you think it means.

  24. TravisM

    Dang. Great movie, but makes for a more interesting article!

  25. Michael

    BA, a problem with the spinning bucket analogy. If you take a bucket, and spin it around you, no water should fall out (until you try to stop it, if you are not careful). This basic theory has been used by Swaggies (swagmen, as in the song), in Australia for a 100 years now (a long time in Australian modern history) to brew their billy (tea). To get the tea leaves to settle in the tin can they brewed it in, they would swing the can around in the circle (usually vertically), not spilling any.

  26. HawaiiArmo

    Princess Bride? It’s a SUpernova 1A, that’s the title pun. Or put it this way, anybody 1A Peanut (the shape of the structure I guess)

  27. Irishman

    PK, look at the linked article, you will see those modelling pics are from there.

  28. Stark

    Michael, you are coreect – assuming the velocity is high enough to keep all the liquid in place and there are no peturbations introduced. The BA’s analogy was right but you need to add a water-proof firecracker into the bucket to get the full effect. When the firecracker goes off you will most certainly get some spillage out of the bucket. Unless you used a very large firecracker that spillage will travel away from you (though at a slower rate than if you had simply dumped to bucket) after it splashes out.

    It’s hard to do but you can demontrate this for yourself with a rock on a string and that bucket of water on a rope – swing them both at the same time and, if you are exceptionally coordinated, you should be able to drop the rock into the bucket. I said it was hard… but I’ve seen similar done (beer bottle on rope and bucket on a rope – it was drunken geek bet).

  29. darius


    You’re making assumptions about the conditions of the bucket. The BA didn’t say you would be holding the bucket straight out, he didn’t say how full it would be, and he didn’t say how fast you would be spinning. You could be holding the bucket upright, it could be filled all the way to the top, and you could be spinning slowly. Almost certainly, some would spill out. There are also various other combinations that would result in a spill.

    Or you could just do what Stark says and add in the dwarf nova effect.

    Are there rocks ahead?

  30. random6655321

    The best part about the Princess Bride quote is that in the movie, it’s a giant asking a dwarf, for which he blows up saying, “Gahh!!!!” lol

    This is a wonderful posting BA.


Discover's Newsletter

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


See More

Collapse bottom bar

Login to your Account

E-mail address:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »