Protoplanetary Disk Evolution: Singles vs. Binaries
Sebastian Daemgen, Ray Jayawardhana, Monika G. Petr-Gotzens, Elliot Meyer
aa r X i v : . [ a s t r o - ph . S R ] A ug Young Stars & Planets Near the SunProceedings IAU Symposium No. 314, 2015J. H. Kastner, B. Stelzer, & S. A. Metchev, eds. c (cid:13) Protoplanetary Disk Evolution:Singles vs. Binaries
Sebastian Daemgen , Ray Jayawardhana , Monika G. Petr-Gotzens ,and Elliot Meyer Department of Astronomy & Astrophysics, University of Toronto, 50 St. George Street,Toronto, ON, Canada M5H 3H4, email: [email protected] Faculty of Science, 4700 Keele Street, Toronto, ON M3J 1P3, Canada European Southern Observatory, Karl-Schwarzschildstr. 2, 85748, Garching, Germany
Abstract.
Based on a large number of observations carried out in the last decade it appears thatthe fraction of stars with protoplanetary disks declines steadily between ∼ ∼
10 Myr.We do, however, know that the multiplicity fraction of star-forming regions can be as high as >
50% and that multiples have reduced disk lifetimes on average. As a consequence, the observedroughly exponential disk decay can be fully attributed neither to single nor binary stars and itsfunctional form may need revision. Observational evidence for a non-exponential decay has beenprovided by Kraus et al. (2012), who statistically correct previous disk frequency measurementsfor the presence of binaries and find agreement with models that feature a constantly high diskfraction up to ∼ . <
100 AU) are significantly reduced compared towider binaries. The frequencies of accretors among single stars and wide binaries appear indis-tinguishable, and are found to be lower than predicted from planet forming disk models governedby viscous evolution and photoevaporation.
Keywords.
Stars: pre-main sequence; Stars: formation; circumstellar matter; binaries: general
1. Introduction
The formation of gas giant planets requires significant amounts of gas and dust to bepresent in the circumstellar environment of a young T Tauri star. The lifetime of pro-toplanetary disks is accordingly an important observable to constrain planet formation.To infer disk lifetimes, a number of previous studies have targeted young star-formingregions to measure the fraction of stars that exhibit either ongoing accretion or hot cir-cumstellar dust or both. These fractions appear to be a strong function of the age of astar-forming region, monotonically decreasing from &
80% to 0% within ∼
10 Myr (e.g.,Jayawardhana et al. 2006; Fedele et al. 2010). The functional shape appears to be fit wellby an exponential decay with a time constant of τ ≈ The effect of stellar binarity on disk evolution.
Sub-mm observations show thatmultiple stars exhibit a much reduced total dust mass compared to undisturbed systems(Harris et al. 2012; Andrews et al. 2013). This can be explained by the dynamical inter-action of a disk with the stellar companion. For wide binaries this leads to a truncation1 S. Daemgen et al.of the outer disk to ∼ τ ≈ . ρ <
100 AU) binary companions are less likely to be orbited by a low-mass ( . Jup ) gasplanet than wider binaries and single stars. One possible interpretation calls for a slow gasplanet-forming process for planets < Jup – too slow to complete in close binaries beforethe disk disperses. More massive planets, in contrast, may form through a rapid processthat completes equally often around single stars and around even closely separated binarycomponents where disk lifetimes are shorter.
The difficulty of identifying single stars.
While above considerations and the factthat sample sizes are typically small paint a noisy but consistent picture of disk evolutionin T Tauri binaries , the evolution of a disk in an undisturbed system like that of a singlestar cannot be easily observed. This is partly due to our lack of knowledge about thetrue multiplicity status of young stars: efficient high-angular resolution imaging searchesfor companions do not reach much below ∼ ′′ ( ∼
10 AU at the distance of the neareststar-forming regions of ∼
100 pc), and few young stars have been subject to radial velocity(RV) monitoring over baselines >
10 yr (equivalent to & ∼
30% or larger, depending on the distanceof the targeted region and the availability of suitable survey data (Kraus et al. 2012;Daemgen et al. 2015, submitted ). The previously derived disk fractions in these samplesaccordingly represent a mixture of binary and single star disk data compromising theirpotential to provide physical information about disk decay.
Disk evolution around single stars.
Applying a statistical correction for undetectedbinarity, Kraus et al. (2012) study the early evolution of single stars and find observa-tional evidence for a disk fraction evolution that may be different from an exponentialdecay. In qualitative agreement with disk evolution models by Alexander & Armitage(2009), disk frequencies stay close to 100% until ∼ ∼ rotoplanetary Disks: Singles vs. Binaries F r a c t i o n o f s t a r s w i t h D i s k s [ % ] Age of Star-forming region [Myr]
Cha I singles
H o t D u st model (Alexander & Armitage 2009)
Fedele et al. 2010
Binaries (Daemgen et al. 2012, 2013) singles (Daemgen et al. 2015, subm. ) Close binaries (25-100 AU)
Wide binaries (100-1000AU)
Close binaries (25-100 AU)
Wide binaries (100-1000AU)
Accre tio n
Figure 1.
Disk frequency of single stars and components of binary stars in the Orion NebulaCluster (1 Myr) and Chamaeleon I (2 Myr).
Left: accretion inferred from Brackett- γ emission. Right: hot circumstellar dust as inferred from H – K and K – L ′ excess (Daemgen et al. 2012,2013, 2015). The blue continuous and dotted lines show exponential decay functions measuredby Fedele et al. (2010, τ accr ∼ . τ dust ∼
2. A coherent analysis of disks in single and multiple stars
We measured the frequency of disks around 52 spatially resolved multiple stars withseparations between ∼ γ (Br γ ) emissionas well as hot inner dust inferred from near-infrared color excess.We compare this binary data set to single stars in Chamaeleon I (Daemgen et al. 2015, submitted ). In order to compile a sample with as little binary contamination as possible,we queried the high-angular resolution imaging survey by Lafreni`ere et al. (2008) forstars without stellar companions between ∼ K s ≈ K -band spectroscopy to assessthe presence of emission at the wavelength of Br γ . Using a Monte Carlo simulation, wesimulated the overall companion distribution of our sample and find that, even afterexcluding all adaptive optics (AO) and RV companions, there remains a ∼
30% chance ofmultiplicity for any star in the sample. Using the results from our Chamaeleon I binarystudy (Daemgen et al. 2013), we correct the measured single star accretion frequency forthe bias introduced by undetected binarity.As these measurements are based on a well-defined set of suitable disk indicators (Br γ ,NIR color excess) and good coverage by AO and RV companion searches, the resultingdisk evolution measurements represent the most robust assessment of the relative abun-dance of disks around the individual components of binaries and single stars to date.
3. Results
Figure 1 shows our assessment of single and binary star disk fraction as a functionof age. Close binaries ( ρ
100 AU) appear to disperse their disks on much smallertimescales than wider binaries, both for our accretion and hot inner disk measurements.We furthermore robustly confirm that the inferred single star accretor fraction ( F =48 +14 − %; Daemgen et al. 2015, submitted ) is about 6 times larger than that of close bi- S. Daemgen et al.naries. The single star accretor fraction appears to be slightly larger but consistent withboth wide binaries and the decay curve found by previous surveys without binary cor-rection (e.g., Fedele et al. 2010). In particular, our new measurement of the single staraccretor fraction is less than 1 σ larger than previous estimates of the accretion frequencyin Chamaeleon I ( ∼ ∼
4. Summary and Conclusions
Single stars are hard or even impossible to identify in most star-forming regions.This is because contamination with binary companions that are inaccessible to today’sobservational methods is on the order of &
30% for a typical star-forming region likeChamaeleon I. As close binaries ( <
100 AU) have been shown to exhibit disk evolutiontimescales that are shorter than those of wide binaries and single stars, typical diskfrequency studies are biased by undetected binary companions. Rather than followingan exponential decay, as suggested by studies without binary correction, the fraction ofsingle stars with disks may stay constant and close to 100% until ∼ . <
100 AU) in the same region. The single star diskfrequency appears to be inconsistent with the very high disk frequency at 2 Myr suggestedby previous models and statistical considerations. Part of the discrepancy between theobservations and models might be due to systematic age uncertainties.More measurements of single and binary star disk frequencies are needed to infer theevolution of undisturbed disks as a function of time and to constrain the qualitative andquantitative evolution of disk material in both truncated and undisturbed disks.
References
Alexander, R. D. & Armitage, P. J. 2009, ApJ, 704, 989Andrews, S. M., Rosenfeld, K. A., Kraus, A. L., & Wilner, D. J. 2013, Apj, 771, 129Artymowicz, P. & Lubow, S. H. 1994, ApJ, 421, 651Bell, C. P. ., Naylor, T., Mayne, N. J., Jeffries, R. D., & Littlefair, S. P. 2013, MNRAS, 434, 806Daemgen, S., Correia, S., & Petr-Gotzens, M. G. 2012, A&A, 540, 46Daemgen, S., Petr-Gotzens, M. G., Correia, S., et al. 2013, A&A, 554, 43Daemgen, S., Meyer, E., Jayawardhana, R., Petr-Gotzens, M. G. 2015, submitted to A&Asubmitted to A&A