Over twenty percent of the planets we’ve found around other stars inhabit binary systems. It’s intriguing to take a close look at these. Most of the planet-bearing binaries are what is known as ‘wide S-types,’ meaning that the companion star orbits the inner star/planet system at a distance of over 100 AU. But take a good look at GJ86b, γ Cephei b and HD41004b. Here we’re looking at three planets in close binary systems with a separation between the component stars of 20 AU or less. That separation raises the eyebrows, for Alpha Centauri A and B form a close binary with a semimajor axis of 23.4 AU.
We have three ongoing planet hunts around the Centauri stars, Debra Fischer’s work being matched by Michel Mayor’s team at La Silla and both complemented by a new search based at Mt. John Observatory in New Zealand. So it may not be long — months, possibly — before we have some word about planets around these stars. Informing all these searches, though, is the issue of gravitational perturbations caused by the proximity of the two stars. Various simulations have shown that planets of Earth mass can exist around stars like this, but recent work has questioned whether planetary embryos can form and remain stable in the first place, a question tackled in a new paper.
Astronomers Ji-Wei Xie, Jian Ge (University of Florida) and Ji-Lin Zhou (Nanjing University, China) have created a series of simulations to estimate how long it takes planetesimals embedded in a protoplanetary disk to accrete into planetary embryos. This is rapidly becoming the key issue. What we know so far is that the mass upper limit of a planet in this system is 2.5 Jupiter masses around Centauri A and 3.5 Jupiter masses around Centauri B. That leaves us with the possibility of lower-mass planets if this system will allow planetary embryos to form. The problem is that most simulations have assumed a disk of such embryos as the starting point and have followed planet development from that point.
From the paper:
…the remaining question is whether these embryos can form. Thébault at al. (2009) recently addressed this problem by analyzing the conditions for planetesimal accretion. They conclude planetary embryos formation through planetesimal accretion seems impossible around α CenB, unless the binary separation was wider in its initial stages. However, their conclusions are limited in the absolutely coplanar case, where the inclination between the gas disk and binary stellar orbit is exactly equal to zero…
Ji-Wei Xie and team reopen the case by extending Thébault’s work to include the effect of binary inclinations. They develop a new model tested through simulation to study whether the zone from 0.5 to 2.5 AU around Centauri B may allow planets to form. The simulations vary the gas-disk density as well as the binary inclination and work with variables like planetesimal mass distribution, impact rate and the fraction of accreting collisions to come up with a conclusion: Planetesimal accretion into planetary embryos takes significantly longer in this binary environment than around single stars, which does not favor the formation of gas giant planets, but the formation of smaller, terrestrial worlds is possible.
Let me quote the concluding paragraph on this:
Our results support recent work by Guedes et al. (2008), which has shown Earth-mass planets can be formed near the habitable zone (0.5-0.9 AU) of α CenB if the disk is initially composed of lunar-mass planetary embryos. The possible accretion zone shown in this paper is roughly between 1-2 AU, which matches well with their planet formation zone (~0.5-2.0 AU)… In addition, at the time of writing this paper, we note a promising result from Payne et al. (2009) that Earth-like planets can also form in the habitable zone of α CenB-like binary systems through outward migration from the inner accretion-unperturbed zone (within ~ 0.7 AU). Therefore, by combining these studies…and our simulations, it is quite possible that a habitable Earth-like planet may be hidden around α CenB.
That’s good news for planet hunters, of course, and the happy fact is that we should be getting solid data to measure these simulations against in the not distant future. The paper is Ji-Wei Xie et al., “Planetesimal Accretion in Binary Systems: Could Planets Form Around α Centauri B?” Astrophysical Journal 708, pp. 1566-1578 (abstract / preprint). For a discussion of Thébault’s work, see this earlier Centauri Dreams story.
Of the three systems you mention, the most extreme is probably Gliese 86. The other two have substantially more unequal mass ratios than Alpha Centauri, so the discs around the primary star would have extended further, providing more material. At Gliese 86 on the other hand, the secondary star is a white dwarf. This means not only was it once the more massive star in the system (the best-fit mass for the progenitor star is about 2 solar masses, over twice the mass of the K-dwarf star Gliese 86 A), it also implies the semimajor axis of the binary system has increased. This makes things very unfavourable for forming giant planets around Gliese 86 A at the time of the system’s formation.
In fact, it might be easier to explain the system by invoking planet formation much later in its history: when the white dwarf progenitor is losing mass during the final stages of its evolution. Some of this matter could have been captured around the now-primary star, such discs have been observed, e.g. around Mira B. The wider separation and more favourable mass ratio, plus the likely increased metallicity of the matter forming the disc could have resulted in much more favourable conditions to form the giant planet.
Such a scenario wouldn’t apply to Alpha Centauri, since none of the stars are white dwarfs, but perhaps at some future stage in the system’s history it may also have more chances at forming planets.
ummm… so, I own property on the moon, and on Mars. But regarding this possible planet around Centauri B, whom would I contact if I wished to make a real estate investment???
:)
-Zen Blade
Major limiting factor: if the disk is initially composed of lunar-mass planetary embryos. Getting from dust grains to this size involves a lot of aggregation. Hence, I remain unconvinced.
Whats the problem with the preprint? My browser(s) cannot open it, some say that the file is corrupted.
Ramond, it opens fine here. Not sure what the problem is — maybe someone else has experienced this and can weigh in?
Planets in close binary systems have already been found even though radial velocity surveys have preferrentially avoided such systems! For example, a gas giant exists in the Gamma Cephei system. According to our best understanding of how planets form, a gas giant is harder to form than a smaller terrestrial type world. So, if a few gas giants have been binary systems, then surely it is not unreasonable to expect that even closer in rocky planets will also be found once we are technologically capable of finding them.
As far as I am concerned, the burden of proof is on those who contend that planet formation in the Centauri system is unlikely given that astronomers have already found planets in several close binary systems. It sure is interesting how these skeptics continue to ignore evidence that puts such an obvious dent in their pessimistic musings. I am with the administrator on this one: I am betting in favor of Centauri planets and I also think Kepler, which is looking at binary systems as well as singles, will find terrestrial planets in close binary systems.
spaceman: I think it is unfair to those who are doing the research to say that they are ignoring these systems. There are several papers studying planet formation in the case of Gamma Cephei (which of these ~20 AU systems has the best-characterised stellar orbit) and these are frequently cited by those studying Alpha Centauri, including the Thébault et al. papers which suggest accretion is difficult. But Alpha Centauri is not Gamma Cephei: the main damaging effect is Alpha Centauri’s more equal mass ratio.
In terms of the outermost stable orbits, Alpha Centauri A and B are more severely-restricted environments than any of the known extrasolar planet hosts, other than Gliese 86 A under the assumption that its planet is indeed primordial and not second-generation. Furthermore the planet population around relatively close binaries (including a wider sample than the few known ~20 AU systems) shows statistically significant differences from the population around single stars and wide binaries. This may perhaps indicate a different mode of formation for the planets. We may be looking at the products of rapid disk collapse to form gas giants rather than accretion (the process thought to form terrestrial-type planets) – such a scenario would be consistent with the observed short disc lifetimes in such systems. Given the apparently rapid formation timescales for giant planets as opposed to terrestrials, the known giant planets around ~20 AU systems may be misleading when it comes to Alpha Centauri.
I personally expect that there are some planets at Alpha Centauri (but pessimistic as to whether an Earth-mass planet located in the HZ will have enough water to be habitable), but the sceptical viewpoint regarding Alpha Centauri planets is not as unreasonable as you are making out.
A correction to the affiliations should be made: Ge is at the University Florida, while Zhou is at Nanjing.
Whoops! Thanks for catching this. I’ll change it right away.