This illustration shows an exoplanet orbiting around two brown dwarfs –- objects bigger than gas-giant planets but too small to be proper stars. Credit: ESO/ M Kornmesser.
The brown dwarf binary system 2M1510 was first discovered in 2018. Located around 120 lightyears from Earth, two brown dwarfs, substellar objects that have masses between gas giant planets and low-mass stars, were seen to eclipse periodically, signifying they were in a short orbit with each other. A third, much more distant, brown dwarf has since been identified, making the system more unusual still.
But recent research has added another layer of complexity. There appears to be a single exoplanet orbiting the central pair. To add to the uniqueness, this exoplanet is not orbiting in a typical plane, but is at 90° to the two central brown dwarfs. The University of Birmingham’s Thomas Baycroft tells ScienceVega about this serendipitous discovery in a very remarkable cosmic system…
ScienceVega: What are brown dwarf stars?
TB: What makes something a star and what makes something a planet isn’t always so clear. In our Solar System, there’s an obvious distinction between our star, the Sun, and the planets in that the Sun is a lot bigger, more massive and it’s brighter. But as we look through the Galaxy, we see objects that range smoothly all the way from a star down to a planet.
A brown dwarf is somewhere between a star and a planet and there are two different ways that you can try to categorise them as star-like or planet-like. The first is whether they have fusion. Typically a star such as the Sun has fusion happening in its core which is powering it and allowing it to emit light. A planet does not. We know the Earth is still hot on the inside from the residual heat from when it was formed and radioactive decay, but it isn’t generating heat from nuclear fusion. A brown dwarf, however, does fuse deuterium in its core (when a hydrogen atom (one proton) combines with a neutron to create a deuterium atom) but it doesn’t have enough mass and enough density in its core to go further. It does undergo a little bit of nuclear fusion, and therefore it does emit light and is a bit brighter than a planet. It is more massive than Jupiter, but not quite as massive or bright as a star. It also won’t have the full mean sequence lifetime and evolve in the same way as a star would.
The other way to look at them is how they form. We think stars form from a cloud of gas and dust collapsing, whereas we think planets form around an existing star, from a disc of material. When we look at brown dwarfs, they seem to be split between more massive one and less massive ones, and fewer in the middle ground. We think the less massive ones probably form like planets, but end up very massive, and the high mass ones form like very light-weight stars; they didn’t quite make it to the mass needed to be a true star. Brown dwarfs are a poorly understood class of objects which are important to study more.
This is an artist’s impression of the exoplanet 2M1510 (AB) b’s unusual orbit around its host stars, a pair of brown dwarfs. The newly discovered planet has a polar orbit, which is perpendicular to the plane in which the two stars are travelling. Polar planets around single stars had been found before, as well as polar discs of gas and dust capable of forming planets around binary stars. But thanks to ESO’s Very Large Telescope (VLT) this is the first time we have strong evidence that such a planet actually exists in a polar orbit around two stars. The two brown dwarfs appear as a single source in the sky, but astronomers know there are two of them because they periodically eclipse each other. Using the UVES spectrograph on the VLT they measured their orbital speed, and noticed that their orbits change over time. After carefully ruling out other explanations, they concluded that the gravitational tug of a planet in a polar orbit was the only way to explain the motion of the brown dwarfs. Credit: ESO/ L Calçada.
ScienceVega: What is so unusual about 2M1510?
TB: What makes this particular system interesting, and what has been interesting for a few years, is that there are two brown dwarfs orbiting each other on a short orbit and they were seen to eclipse about five years ago. One of them passed in front of the other and it was observed from the ground that the light that was being emitted had a clear dip as one brown dwarf obscured the other. We only know of two young pairs of brown dwarfs that do this.
They are interesting in their own way, because they help us understand how brown dwarfs work. When you see them eclipse, you can get measurements of how big they are (what their radius is) and then comparing that to their mass we can start to understand how these objects form, cool and contract, and how are they expected to evolve, which is an interesting question in science at the moment.
This system has these two brown dwarfs that are eclipsing, but there is a third brown dwarf very far away. That brown brand dwarf is at least at 250 astronomical units (one AU is the distance from the Earth to the Sun (approximately 93 million miles)) from the central pair. That means we have a triple brown dwarf system, which makes it even more interesting; how do you form a system with so many objects that are so similar?
That was the question going into our project, but we ended up finding evidence for a planet in this system as well in a peculiar orbit. That was quite serendipitous. We weren’t expecting to find it, we were just trying to study the system and have found something potentially very remarkable.
ScienceVega: How rare are polar planets that orbit at 90° angles to their stars rather than in the more normal plane?
TB: Around single stars, we know of polar planets where the star is spinning in one direction and the planets are orbiting in a polar configuration. They are reasonably rare, but it’s also quite difficult to make the required measurements. When you see a star, you don’t really know the direction it’s rotating in, and getting a good constraint is not easy. From the ones where it has been possible to measure it seems to be that the majority of planets are in the same or similar orbital plane as the spin of their star. Then there is also a small population that are close to a polar orbit. It’s not uncommon but it’s definitely the rarer configuration.
ScienceVega: Do we know why there’s a difference between planar and polar orbits?
TB: In the simple model of planet formation, you have a star that forms and there is a cloud of gas and dust around it that ends up collapsing into a disc. Because of the angular momentum, that means everything slowly flattens out. Typically, as with our Solar System, we think that would lead to the planets all being in the same plane around the star that’s also rotating in that same plane.
To make a planet become polar there needs to be some other sort of phenomenon that is disruptive. It could be another star nearby that causes some sort of excitation of three body dynamics that moves things around, or even just a nearby star passing by. For example two stars that formed recently and one of them just happens to pass nearby in its orbit through the Galaxy, which could stir things up. You could have multiple planets in the same system, and perhaps two giant planets have an encounter where they get too close to each other and interferes with each other’s orbits.
The typical picture if not very much happens, is you would expect them to stay coplanar. You require something more, maybe even something dramatic to happen for a planet to become polar.
This image, taken in visible light, shows 2M1510 AB, a pair of brown dwarfs orbiting each other. The two brown dwarfs, A and B, are seen as a single source in this image, but we know there are two of them because they periodically eclipse each other. When monitoring their orbits, astronomers found perturbations that can only be explained by the gravitational tug of an exoplanet circling both brown dwarfs in a perpendicular orbit. This system contains a third brown dwarf, 2M1510 C, which is located too far away to be responsible for these perturbations. Credit: DESI Legacy Survey/ D Lang (Perimeter Institute).
ScienceVega: How common are these two-star systems with planets orbiting both stars?
TB: Multiple star systems are very common. Most stars in the Galaxy are in binaries or triples or higher order systems. However, it’s not so easy to quantify the rate of planets in those systems, but we have evidence that they’re pretty common.
There are two main types: you can have two stars that are quite far apart, and you have a planet orbiting around one of them. We know of a couple of hundred planets in this configuration. Bearing in mind we know of about 5,000 to 6,000 planets total, 400 or so around a single star in a wide binary is a reasonably high number.
A planet around a close binary, where you have the two stars on the inside and the planet further out, which is what I’m most interested in, are called circumbinary planets and perhaps a dozen or more have been found. Some are confirmed and some are candidates.
Despite that, we actually think they’re quite common. In terms of gas giant planets like Jupiter and Saturn around binary stars, they are just as common as those around single stars. The difficulty is that it is harder for us to detect them, so we haven’t quite found as many at the moment. But based on the statistics, we think they’re about as common.
ScienceVega: This is a first where an exoplanet is orbiting both stars at the 90° polar direction – do you think they are simply very rare, or have our detecting capabilities improved and we will find lots more?
TB: I think it’s both. Firstly, this is evidence of such a planet that we will need to keep following up to make sure that it’s real. It is interesting to find evidence of such planets, because they’re existence has been theorised. There are polar planets around single stars, so clearly there are mechanisms that allow the planets to get shifted into such orbits, and especially around binary stars it’s a very stable configuration to be in such polar orbits. People have been predicting that they should or at least could exist.
In this case, we found evidence of something unusual going on with the binary orbit, this retrograde apsidal precession, as it is known. We went through and considered what various options exist that could cause such a such a motion and effectively ruled out each one in turn such that it can only really be a polar planet.
But we can’t actually say anything about the mass of the planet. It could be a low mass planet, very close to the binary, or it could be a high mass planet further away, and it could go range all the way to a low-mass brown dwarf, perhaps 15 to 20 times the mass of Jupiter. Anything higher than that and we would expect to have seen its light in our spectra. It could be quite far away from the binary and still be causing the same influence as something of lower mass and nearby.
We’re calling it a planet candidate at the moment, but we we’re planning to get more data and keep observing firstly to confirm the measurement we made and try to also get independent confirmation of the planet from a different type of data.
This is a nice thing about how science works in that you posit a hypothesis and you can say that it’s not certain, but then we can go and test it again and make sure.
As to whether they are rare or we’ll find more, from what we can see of planets around single stars, there’s lots of coplanar ones and a few polar ones. So if that’s similar for circumbinary planets, we should expect that they’re rarer, but I do think we will be able to find more, just maybe not as many. It has been difficult to do so with the methods that have been used so far, but if we try with some targeted efforts to look specifically for these kinds of signatures, hopefully we’ll find more. That is what I’m hoping to do eventually.
The planet, called TOI 1338 b, is around 6.9 times larger than Earth, or between the sizes of Neptune and Saturn. It lies in a system 1,300 light-years away in the constellation Pictor. The stars in the system make an eclipsing binary, which occurs when the stellar companions circle each other in our plane of view. One is about 10% more massive than our Sun, while the other is cooler, dimmer and only one-third the Sun’s mass. TOI 1338 b’s transits are irregular, between every 93 and 95 days, and vary in depth and duration thanks to the orbital motion of its stars. TESS only sees the transits crossing the larger star — the transits of the smaller star are too faint to detect. Its orbit is stable for at least the next 10 million years. The orbit’s angle to us, however, changes enough that the planet transit will cease after November 2023 and resume eight years later. Credit: NASA/Goddard Space Flight Center, Public domain, via Wikimedia Commons.
ScienceVega: Can you explain the processes by which you came to the discovery?
TB: The initial discovery of the binary eclipses was done with the SPECULOOS (Search for habitable Planets EClipsing ULtra-cOOl Stars) telescopes in 2017, and the paper was out in 2020. At that time that was a serendipitous discovery. They were observing this system, thinking it was a normal star, but then it turned out they saw an eclipse and found that it was too lower mass, which meant they were looking at brown dwarfs. At the time they collected some data with UVES (Ultraviolet and Visual Echelle Spectrograph) which is a spectrograph instrument on the ESO’s Very Large Telescope, and from the spectrograph data, you can measure the velocity of the stars in the line of sight. Combining that with the eclipse, we can characterise the orbit of the binary star, so we can measure the masses and have a sense of the radii based on the eclipse, which was done at the time.
Then a few years ago, we collected some more data with the same instrument. The idea was to get a better measurement of the masses, try and understand the system better, and be able to recover more eclipses. There has only been one eclipse observed so far, the one in 2017. We’ve tried a handful of times to search for the eclipse again, but without knowing exactly when it’s going to happen you have to keep observing at night, but maybe it happened during the day when you can’t observe it from the SPECULOOS telescopes. We are hoping to try and recover more eclipses, to get a better understanding of the orbit, and study the system as an interesting pair of brown dwarfs.
When analysing the data, we see that the elliptical orbit of the binary is changing orientation, which is called apsidal precession; effectively the elliptical orbit is rotating slowly in time. This type of motion was seen in our own Solar System: the apsidal precession of Mercury was a pretty big deal at the start of the 20th Century. It was one of the few first big successes of Einstein’s theory of general relativity predicting the excess motion that was observed in Mercury’s orbit wasn’t because of an extra planet we didn’t know about, but was just a consequence of gravity.
This is the same extra motion where we’re seeing this ellipse changing orientation, but it’s going backwards. It is going in the opposite direction than it would from general relativity and most other mechanisms. One of the few things that can cause this retrograde precession is if you have a third body on a polar orbit. That is why we think we have found evidence of such a such a companion.
ScienceVega: What you want to do next given these findings of this particular bit of research?
TB: In my future research I want to find more of these planets or more evidence of polar circumbinary planets. Once we’ve established that something exists in nature, the next interesting question scientifically is how common are they? I want to try and be able to do a thorough comparative study and get a sense of how common are the polar ones relative to the coplanar circumbinary planets. This could involve measuring the precession rate on a large sample of binary stars, and seeing how many are negative versus how many are positive, as a first step towards working out the relative occurrence or frequency of such polar circumbinary planets.
I also want to look at systems around single stars where you have two planets, but they have different inclinations, seeing how they are related, whether they form in similar or very different ways and getting an understanding of planetary systems with any number of stars in a three-dimensional way.
Illustration showing a hypothetical office overlooking the Paranal Observatory in Chile, with the European Southern Observatory’s VLT visible with its laser on the hill, and the four small SPECULOOS telescopes nearer the foreground. Copyright: University of Birmingham / Amanda Smith.
The link to the original paper can be found here.
