Interview: Ian O'Neill chats with Dr. Richard Barry, an astrophysicist at NASA's Goddard Space Flight Center, about the recurrent nova, RS Ophiuchi.

The scoop: Ian O'Neill chats with Dr. Richard Barry, an astrophysicist at NASA's Goddard Space Flight Center, about a very special binary star system called RS Ophiuchi where a recurrent nova generates a fascinating spiral dust cloud in its wake.

Richard Barry: Ready to go!

Ian O'Neill: Hi Rich!

Rich: I sent you some documents and artwork to go with our discussion. I also sent a movie. I hope those will be helpful.

Ian: Great stuff, I'm having a look now, thanks for that!

I was thinking about asking some general questions about recurring variables and then focusing on your research into RS Ophiuchi. I find that binary system fascinating!

Rich: Good. Let's talk about recurrent novae in general then. s

Ian: I actually spoke with the development director of the AAVSO and there was quite a bit of overlap with their variable star observations.

Can recurring novae be called "recurring variables" or is that a mash-up of terms?

Rich: :) Yes, that is a bit of a mash-up. Recurrent novae are a kind of variable star.

Ian: Hehe, ok, made a note of that now. Thanks :)

Rich: So, novae, in the most general sense, are caused by the accretion of hydrogen onto the surface of a small, high density stellar corpse from a companion star.

Ian: So are recurrent novae exclusive to binary star systems?

Rich: Yes, although there could possibly be other mechanisms. The model that holds the most water at present is one in which you have a stellar cinder and another star in orbit about their common center of mass. Material flows from the companion (usually termed the secondary) onto the surface of the primary - often a white dwarf.

Ian: So how many binary systems are there that we know of that have this special relationship? I haven't heard much about recurrent novae before now.

Rich: Well, to clarify, the best model we have, that I alluded to before, indicates that almost all novae should be recurrent novae. Most recurrent novae will erupt on exceedingly long intervals - from tens of thousands to hundreds of thousands of years! There is another class of novae that have erupted more than once in recorded history. We call these 'recurrent.'

Ian: Ah, I see. So the nova itself occurs near the surface of the white dwarf partner? Do we know how the nova is triggered? Is there like a 'critical mass' of hydrogen that needs to accumulate?

Rich: That is a very good question! A nova likely occurs on or near the surface of the white dwarf. (Again, there are other possibilities rather than white dwarf - but we will just stick with white dwarf for simplicity.) Let me start from the beginning and tell you how these are theorized to occur.

Ian: Great, that would be very good to know (it might help with the next big question I have about white dwarfs!)

Rich: All main sequence stars -- stars that are in the hydrogen-burning part of their lives -- and stars that are very old shed lots of matter. The matter flies off of them effectively radially. If the star is a member of a binary and the companion is a white dwarf, some of the matter from the secondary will be captured. This can occur through a process called 'wind capture' or 'Roche lobe overflow.'

So, the matter falls towards the white dwarf (WD) and captured in its gravity well. The material accumulates on the surface of the WD until its density and temperature are so great that it detonates in a surface thermonuclear explosion.

Ian: So how close are the stars in the binary? I'm looking at this stunning piece of artwork you sent me and the MS star seems awfully close to the white dwarf... no wonder the WD grabs a lot of gas...

Rich: The amount of mass -- and here is the answer to your question -- depends critically on the mass of the WD. The greater the mass of the WD, the less mass accumulation required to form a nova eruption.

The distance between the companion stars is another critical element.

The greater the distance between the companions, in general, the longer time it will take for the WD to accumulate the correct amount of mass to erupt.

In the picture I have presented, which was commissioned by NASA and the Keck Observatory for some work we did on a particular recurrent nova, the distance between the companions is about 1.7 AU -- or about 1.7 times the distance between the Earth and the Sun.

Ian: Oh cool, so the stellar distance almost acts as a nova timer? The greater the distance, the longer the time between explosions...

Rich: Don't let me forget! I can connect recurrent novae to supernovae!supernovae

Ian: That was going to be my next question! WDs are behind Type 1a supernovae.... WDs are behind these recurrent novae... what's the difference between the two? I'm assuming Type 1a supernovae only have to happen once...

Rich: Yes, stellar distance is one of the elements that sets the amount of time between outbursts, but the mass of the WD primary is of greater importance.

Ooops. I'm talking over you. I will now answer your very last question.

Ian: It's OK, this is a very exciting topic! I have lots of questions.

Rich: Okay. Deep breath. I just love this stuff!

Ian: I think I do too! It makes my previous coronal loop research seem very pedestrian :)

Rich: WDs are the drivers for most supernovae and most recurrent novae as far as we know. So, for recurrent novae, theory tells us that there is a mechanism that can allow the WD to gradually, secularly, accumulate mass through multiple outbursts. This appears to be the case for the recurrent nova 'RS Ophiuchi' which is the subject of the artwork I sent you.

As the WD slowly accumulates mass over the aeons it may approach a very important critical mass called the Chandrasekhar limit. This is the mass at which the object can no longer mechanically support itself with electron degeneracy pressure. The WD in the nova RS Ophiuchi, for example, is thought to have a mass somewhere around 1.35 - 1.37 times the mass of our sun. The Chandrasekhar mss limit is about 1.44 times the mass of our sun if the WD is not spinning. This number may be somewhat higher if the WD is spinning.

Ian: Ah, now I see. But I suppose the biggest question that I've just thought of... how big are these novae explosions? I'm assuming they're not just a stellar flaring event! m

Rich: That is a great question. Novae, as I have described them, are thermonuclear runaway reactions within the surface layers of the WD. The eruptions are vigorous enough to eject most of the layer of material accumulated from the secondary and are thought to entrain some of the surface material of the WD thus enriching the interstellar medium with some heavier elements.

The important point here is, that in a typical nova event, the WD is not blown up but exists from event to event. This is shown in the artwork. In this representation, the companion star is a Red Giant (RG) - a star that has exhausted most of its hydrogen and has expanded in size. This particular RG has a very massive wind. What is depicted in the art is what we think the system looks like about 3.7 days after the surface explosion on the WD.

You may wish to ask me about the spiral dust pattern... ;) spiral

Ian: So, what about the spiral dust pattern? I feel compelled to ask you about that :) pattern

Rich: In the event that the WD has exceeded the Chandrasekhar mass limit as a result of mass accumulation from its companion, the WD will collapse and be completely destroyed. This enormous explosion occurs at about the same mass for all WD (depending on spin rate) so, as a result, they form an important part of the cosmic distance ladder. cosmic

Ah, yes. I'm very glad that you asked me about the spiral dust pattern. That is brand new theory you are looking at!

Ian: Is that a result from the binary orbit? Or... and I'm guessing here... some kind of pulsing during the nova?

And what elements are found in the dust? Is it as enriched with heavy elements like a supernova remnant?

Rich: You are nearly right! As I mentioned before, RN enrich the interstellar medium by blowing off surface layers of material. In the past (before 2007) it was thought that the way dust is created by RN is that the blast from the event compresses the immediate environment about the binary causing condensation of dust.

For dust creation, you must have a critical density of material so that three-body collisions can occur - two to stick together and one to carry away excess momentum.

Well, we happened to be commissioning the Keck Interferometer Nuller (KIN) when the nova RS Ophiuchi occurred. The KIN uses the combined light from the two largest optical telescope in the world (each of which is 10m) to form the equivalent of an 85m telescope.

We use the interference of the light from the two telescopes to form a fringe pattern - of dark and bright fringes. We then place one of the dark fringes on the brightest part of the object - in this case, the WD in eruption - and can look at the faint material in the immediate vicinity. . This was completely invisible to science ... until now! So, we accumulated 20 minutes of data from the KIN on the nova and in the data we found that there was silicate dust a mere 17 AU from the central bright source! What this meant to us is that the old theory simply did not hold up in the light of this new data. The old theory told us that that dust should not have been there at all. It should have long since been blown away by the wind from the recurring nova! Ian: Wow, that must have been thrilling to see. Was it a surprise discovery or had the dust been predicted before? Rich: This was a complete surprise to all parties. The silicate dust had been unexpected. To form the dust, as I mentioned, you must have enough density of material and it has to be at the right temperature. Using these observations, we have now formulated a new theory of how silicate dust is created by recurring novae! ! Ian: So is it fair to say that RS Ophiuchi has been key to recurrent novae research? Sounds like a fascinating system! Rich: Yes! What you see in the artwork is an Archemidian dust spiral caused by the massive WD plowing through the massive wind from the RG. Then, every 20 years or so, mass that has accumulated on the WD from the RG explodes and completely clears the beautiful spiral! Ian: Absolutely stunning. Rich: Right now, as we 'speak' the spiral should be reforming. At some point in the near future, we may even be able to see the spiral in reflected light using Hubble or possibly JWST. Ian: Actual optical images? Now that would be a great sight Rich: Yes, very beautiful and exciting. This will drive a stake in the old theory! To answer your previous question, recurring novae do enrich the interstellar medium with lighter elements. When they go supernova, THEN they produce the really massive elements. You need a very strong explosion to do that. . Ian: So am I correct in saying that recurrent novae can be a precursor to an eventual supernova? Could RS Ophiuchi (for example) ever get completely destroyed by a supernova? Or is the white dwarf simply too small? Rich: Yes, you are correct. RS Ophiuchi is likely to go supernova at some point in the future. The reason that it erupts every 20 years or so is because the WD is so very near the Chandrasekhar mass limit. It is really wonderful that we have this great laboratory for the study of very dense matter and nucleosynthesis right in our own back yard! Ian: RS Ophiuchi is possibly the most exciting system I have ever heard of! I hope there isn't any planetary systems around either of the binary stars... or in neighboring star systems for that matter. They would have a very bad day (every 20 years or so). Rich: Yes! It is also a good thing that by 'back yard' I mean astronomically speaking. The system is 1.4 [+0.6, -0.2] Kpc away from us. My colleagues and I wrote an entire ten page paper on just the distance to it. It is surprisingly tricky to figure out the exact distance to novae. Ian: Finally, we're in the clear if that thing decides to blow as a supernova then..? Rich: Oh, yes. No worries. We made sure of that. Ian: Great, I'll let everyone know about that good news... a lot of people were freaking out about WR 104, they don't need to worry about RS Ophiuchi as well. Well, thank you so much for chatting with me! Rich: Great, Ian. Lots of fun!