ROXs 42B - Young binary M dwarf with substellar companion(s?)

Direct Imaging and Spectroscopy of a Candidate Companion Below/Near the Deuterium-Burning Limit In The Young Binary Star System, ROXs 42B
http://arxiv.org/abs/1310.4825

Quote :

We present near-infrared high-contrast imaging photometry and integral field spectroscopy of ROXs 42B, a binary M0 star member of the 1--3 Myr-old ρ Ophiuchus star-forming region from data collected over a 7 year span. Each data set reveals a faint companion (hereafter ROXs 42Bb) located ∼ 1.16" (rproj ≈ 150 AU) from the primaries. ROXs 42Bb's astrometry is inconsistent with a background star but consistent with a bound companion, possibly one with detected orbital motion. The most recent data set reveals a second candidate companion at ∼ 0.5" of roughly equal brightness. ROXs 42Bb's H and Ks band photometry is similar to dusty/cloudy young late M/early L dwarfs with masses at or below the deuterium-burning limit. K-band VLT/SINFONI spectroscopy shows ROXs 42Bb to be a cool substellar object (M8--L0; Teff ≈ 1800--2600 K), not a background dwarf star, with a spectral shape indicative of young, low surface gravity companions with masses at/below the deuterium-burning limit. We estimate ROXs 42Bb's mass to be 6--15 MJ, either below the deuterium burning limit and thus planet mass or straddling the deuterium-burning limit nominally separating planet-mass companions from other substellar objects. Given ROXs 42Bb's mass with respect to the primaries and its projected separation, it may represent the lowest mass objects formed like binary stars or a class of planet-mass objects formed by protostellar disk fragmentation/disk instability, the latter slightly blurring the distinction between non-deuterium burning planets like HR 8799 bcde and low-mass, deuterium-burning brown dwarfs.

Also a second candidate companion is imaged in H-band but it needs confirmation.

And did anyone notice arXiv renders LaTeX now in the abstract text?

The paper is vague about host star's characterization. ROXS 42 appears to be a triple, perhaps quadruple system configured this way (data extrapolated from Barsony et al. (2003) and references therein):
 
ROXs A-B: a = 56 milliarcsec = ca. 7.6 AUs
 
ROXs 42 AB-C: a = 0.157 arcsec = ca. 23 AUs
 
ROXs 42 Ca-Cb: a = 0.00195 arcsec (=1.95 milliarcsec) = ca. 0.26 AUs (with P_orb=39 days).
 
"ROXs 42 b": a = 150 AUs
 
If so the substellar companion is neither circumbinary or circumternary but circumquaternary!?
 
It could be labelled either as ROXs 42 (ABC)b or as ROXs 42 (ABCaCb)b,,.

It's not a star system, its... a cluster

Edit: the discovery paper talk about ROXs 42 B being a binary star, and the planet must orbit this binary in order to be names ROXs 42 Bb. So ROXs 42 A + B must be a wide binary! Not a close one like in your paper Edasich.

Very interesting finds Edasich, though SIMBAD suggests ROXs 42A and ROXs 42B are separate, unrelated? systems. I'll dig around and see if I can find anything.

Google scholar automaticaly correct ROXs to rows, enabiling the search...

Not really a recent reference, but quoting from Ghez et al. (1993):
 

Quote :

Mathieu et al. (1988) detected both a spectroscopic binary companion in both 162814-2427 (ROXs 42C) and 155913-2233: thus both systems consist of at least three stars.

 
 
Mathieu et al. (...) actually dates back 1989 and ROXs 42 C itself appears a spectroscopic, highly eccentric (e=0.48) binary system made up of similar stars (M₁/M₂=1.09 or M₂/M₁= 0.92).

So... let me see if I get this straight, ROXs 42 has three major components consisting of five? stars:
ROXs 42A is a K-type star.
ROXs 42B is a binary M0 dwarf and host to a circumbinary planet.
ROXs 42C is a binary star as well?

Edit: Stalker's link suggests ROXs 42A is a spectroscopic binary as well. Six stars? Three binary components?

Not really. ROXs 42A-B should form the inner binary system (K7+M0). ROXs 42 C (K6) is a tertiary star orbiting the inner binary and is in turn binary itself made up of similar stars (so K6+K7). "ROXs 42 b" seems to revolve around the all of them. The binary separations are given in Barsony's paper in 2nd post.
 
However from today's paper:
 

Quote :

Lunar occulation measurements showed it to be a close, nearly equal brightness binary (r  0.′′05) (Simon et al. 1995). Sensitive optical to (sub)millimeter data reveal no clear evidence for circumstellar gas or dust (e.g. Cieza et al. 2007; Andrews and Williams 2007).

 
 
Citing the same reference Stalker has posted here. Assuming 1'' = 135 AUs as in the picture Sirius posted and doing the math you get the aforementioned separations which match especially accounting CaCb system, where a separation of 0.00195'' is given at orbital period of 39.5 days. This translates in a = 0.26 AUs for a couple of 0.8-0.7 M_Sol stars.
 
Assuming same reference you get 7.5-7.6 AUs for A-B inner binary's separation (a = 0.05') and ca. 23 AUs for AB-(CaCb) separation (a=0.157'').

In short ROXs 42 is (supposed to be) quadruple.

But the planet il ROXs 42Bb is orbiting ROXs 42B, otherwise it would be ROXs 42ABb or ROXs 42 b

Stalker wrote:

But the planet il ROXs 42Bb is orbiting ROXs 42B, otherwise it would be ROXs 42ABb or ROXs 42 b

 
For me it should be ROXs 42 (ABC)b.
 
On the other hand if EPE is going to include this in planet list it should not amount to 1,000 planets but 998, since there are two double entries which have not been merged (PH1= KIC 4862625 b and Kappa And b to be removed).

For you, it should be ROXs 42 (ABC)b. But seriously, I think the authors are not stupid^^

Let's not sour this discussion with unnecessary vitriol.

Quote :

On the other hand if EPE is going to include this in planet list it should not amount to 1,000 planets but 998, since there are two double entries which have not been merged (PH1= KIC 4862625 b and Kappa And b to be removed).

EPE's erroneous entries go deeper than that. The addition of Kepler-56 d is a recent example - it's only an RV trend at this point and could just as easily be a star.

Quote :

For you, it should be ROXs 42 (ABC)b. But seriously, I think the authors are not stupid^^

 
 
Paper aimed to announce discovery of substellar/planetary companion. About the stellar system:
 

Quote :

We leave these issues, a more detailed investigation of the host star properties, further analysis of the atmosphere of ROXs 42Bb, and follow-up of the second candidate companion to future studies.

 
 
Host star's characterization was not primary aim of the research. Fothcoming papers are likely going to solve the conundrum.
 
About stellar configuration, here is a quick Celestia model to explain what I mean:
 

Hum, may be. But why the authors named it "Bb"? I'm sorry, but for me it's important to explain this choice! And it's said that ROXs B is a binary separed by 56mas (~7.6 AU) itself in my paper!

Binary designation (especially in case of multiple young stars) may vary and letters A, B, C often do not reflect exact hierarchy.

The naming of circumbinary planets have not been particularly consistent throughout the field. I am not sure I would look so much into how the planet's designation is written.

I added this system to my website with your hierarchy but i'm waiting something more "official". Look at it:

http://extrasolar.fr/index.php?CONTENU=catalogue&star_url=ROXs_42

I named the planet just "ROXs 42 b".

Well this is how SIMBAD has them labelled.



ROXs 42A
ROXs 42B
ROXs 42C

If this is correct, it would seem to imply the planet orbits the B pair only, and the A and C pairs are not orbited by the planet.

Meh... Someone must sent an e-mail to the autors.

To expand on what Sirius shows, the designation does not exactly imply that three stars are bound - the ROXs designation originates from Bouvier and Appenzeller (1992), meaning (the optical counterpart to a-) Rho Oph X-ray source, so the A, B, C, etc. designations refer to individual visual objects at the location of an X-ray object. All three visual components have the proper motions of ρ Oph members, but they do not strictly form a bound system.

Amending what Edasich wrote using Barsony et al., "A-B" refers to Bab (0.056", which is what is being referred to in the titular paper), "AB-C" refers to Cab, and "Ca-Cb" refers to the whichever component of Cab is the SB2 found by Mathieu et al. Therefore, ROXs 42B has only two stellar components.


ANYWAY, reeling this discussion right back to the paper at hand, this is a pretty impressive discovery! The ρ Oph complex is one of the youngest groups of stars that are still relatively easy to study, so finding a… planet? in there is rather useful. I hesitate to refer to it as a planet because it still seems difficult to not form such an object in the same way as a star or BD, but as this is a fairly wide binary the effects of orbital dynamics on the protoplanetary disk would probably have been non-negligible, so maybe standard arguments aren't entirely correct in this case. Given the candidate c is closer in (~10 times the star-star distance), if it is real then it will be a more useful dynamical test of formation mechanisms, though the previous case of Gliese 758 must be remembered.

So presumably if you want to refer to the individual objects, they would be Ba1, Ba2 and Bb?

Which candidate "c" do you mean? Didn't find a reference to an additional planet candidate in ROXs 42B...

Lazarus wrote:

Which candidate "c" do you mean? Didn't find a reference to an additional planet candidate in ROXs 42B...

The planet discovery paper mentions it briefly,

Quote :

Furthermore, the H-band 2011 data set reveal what could be a second candidate at r  0.′′5 ( 65 AU), which has roughly the same brightness as ROXs 42Bb. The data quality in the other epochs is not sufficient to recover this object.

H-band image.

Added to EPE:
 
http://exoplanet.eu/catalog/roxs_42b_b/
 
Now there are these two 20 Jupiter masses objects to make the list complete, besides GSC 08047-00232 b and some more low-mass substellar objects recently treated as possible imaged exoplanets in Neuhäuser & Schmidt (2012), Patience et al. (2012) and more recently in Bonnefoy et al. (2013):
 
http://exoplanetarchive.ipac.caltech.edu/cgi-bin/DisplayOverview/nph-DisplayOverview?objname=HD+203030+b&type=CONFIRMED_PLANET
 
http://exoplanetarchive.ipac.caltech.edu/cgi-bin/DisplayOverview/nph-DisplayOverview?objname=DENIS-P%20J082303.1-491201%20b