The VLA Nascent Disk and Multiplicity Survey: First Look at Resolved Candidate Disks around Class 0 and I Protostars in the Perseus Molecular Cloud

0.41 MB·

by  Dominique M. Segura-Cox

Accepted by ApJL
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
THE VLA NASCENT DISK AND MULTIPLICITY SURVEY: FIRST LOOK AT RESOLVED CANDIDATE
DISKS AROUND CLASS 0 AND I PROTOSTARS IN THE PERSEUS MOLECULAR CLOUD
Dominique M. Segura-Cox1, Robert J. Harris1, John J. Tobin2, Leslie W. Looney1, Zhi-Yun Li3, Claire
Chandler4, Kaitlin Kratter5, Michael M. Dunham6, Sarah Sadavoy7, Laura Perez4, Carl Melis8
Accepted by ApJL
ABSTRACT
6
We present the first dust emission results toward a sample of seven protostellar disk candidates
1
around Class 0 and I sources in the Perseus molecular cloud from the VLA Nascent Disk and Mul-
0
tiplicity (VANDAM) survey with ∼0.05′′ or 12 AU resolution. To examine the surface brightness
2
profiles of these sources, we fit the Ka-band 8 mm dust-continuum data in the u,v-plane to a simple,
n
parametrized model based on the Shakura-Sunyaev disk model. The candidate disks are well-fit by
a
a model with a disk-shaped profile and have masses consistent with known Class 0 and I disks. The
J
inner-disksurfacedensitiesofthe VANDAMcandidatediskshaveshallowerdensityprofilescompared
2 to disks around more evolved Class II systems. The best-fit model radii of the seven early-result
1
candidate disks are R > 10 AU; at 8 mm, the radii reflect lower limits on the disk size since dust
c

continuum emission is tied to grainsize and large grains radially drift inwards. These relatively large
]
R disks,ifconfirmedkinematically,areinconsistentwiththeoreticalmodelswherethedisksizeislimited
by strong magnetic braking to < 10 AU at early times.
S
Subject headings: protoplanetary disks — stars: protostars
.
h
p
- 1. INTRODUCTION tions (e.g., Machida et al. 2014). The protostellar stage
o
representsthestagewiththelargestmassreservoiravail-
r Disks around young protostars are intrinsically linked
t to planet formation, binary formation, and protostellar able to form disks, therefore understanding the proper-
s ties of disks at early epochs is crucial to determine the
a massaccretion(Williams & Cieza2011;Armitage2011).
formation mechanism behind these structures.
[ In spite of their importance in star formation, when
Protostellar disks, however, are difficult to observe.
and how disks form–andtheir properties at early times–
1 Class 0 protostars and disks are so enshrouded that
are poorly constrained. Hence, whether large and mas-
v &90% of their millimeter emission comes from the enve-
sive disks can form in young protostars (i.e. Class 0/I
0 lope (Looney et al. 2000). The three Class 0 protostars
stages, Andr´e et al. 2000; Dunham et al. 2014) is sub-
4 thathavebeenobservedwithenoughresolutionandsen-
ject to debate (e.g., Jørgensenet al. 2009; Chiang et al.
0 sitivitytodeterminetheirdiskpropertiesanddetectKe-
3 2008;Maury et al.2010,respectively). Theoreticalstud-
plerian rotation(L1527, VLA 1623,and HH212) have R
0 ies demonstrate magnetic fields can affect the forma-
> 30 AU disks, larger than expected from strong mag-
. tion timescales and properties of disks. In particular,
1 netic braking models (Ohashi et al. 2014; Tobin et al.
strong magnetic fields can remove angular momentum
0 2012; Murillo et al. 2013; Codella et al. 2014). Observa-
from a collapsing envelope via magnetic braking, re-
6 tional limitations prevent smaller disks from being de-
ducing forming disks to R < 10 AU (e.g., Mellon & Li
1 tected; these Class 0 disks may not represent typical
2008; Dapp & Basu 2010; Machida et al. 2011; Li et al.
: disks at this stage of evolution. Class I protostars are
v 2011; Dapp et al. 2012). Complicating the issue, sev-
less embedded and have cleared enough of their mass
i eralmechanisms can lessen the effects of magnetic brak-
X reservoir that more disks have been detected than in
ing, leading to larger disks: misalignment between
r the rotation axis and magnetic field of the system Class0systems(e.g.,Harsono et al.2014),thoughnotas
a many disks have been revealed as in more-evolvedClass
(e.g., Joos et al. 2012; Li et al. 2013), turbulence (e.g.,
IIsources(e.g.,Andrews et al.2009,2010). Withoutde-
Seifried et al. 2013; Joos et al. 2013), and initial condi-
tectionandstudyofmoredisksaroundtheyoungestpro-
tostellarsystems,wecannotcharacterizeearlyconditions
1DepartmentofAstronomy,UniversityofIllinois,Urbana,IL of mass accretion onto the central protostar and planet
61801, USA;[email protected] formation because evolutionary mechanisms can change
2LeidenObservatory,LeidenUniversity,P.O.Box9513,2000-
disk properties by the Class II phase (Williams & Cieza
RALeiden,TheNetherlands
3Department of Astronomy, University of Virginia, Char- 2011).
lottesville,VA22903, USA To characterize the properties of the youngest disks
4NationalRadioAstronomyObservatory,Socorro,NM87801, and binaries, we are using the Karl G. Jansky Very
USA
5Steward Observatory, University of Arizona, Tucson, AZ Large Array (VLA) to conduct the VLA Nascent Disk
85721, USA and Multiplicity (VANDAM) continuum survey at λ ∼
6Harvard-Smithsonian Center for Astrophysics, Cambridge, 8,10,40,64 mm toward all identified protostars in the
MA02138, USA Perseus molecular cloud (Tobin et al. 2015a). The
7Max-Planck-Institut fu¨r Astronomie, D-69117 Heidelberg, Perseus molecular cloud is relatively close (d ∼ 230 pc;
Germany
8Center for Astrophysics and Space Sciences, University of Hirota et al. 2008, 2011) and has a significant number
California,SanDiego,CA92093,USA of Class 0 and I protostars (43 and 37 systems, respec-
2 Segura-Cox et al.
VLA 8mm Data Model Residual 0.200 VLA 8mm Data Model Residual 0.200
0.175 0.175
) )
0.150am 0.150am
0.125be 0.125be
0.100Jy/ 0.100Jy/
m m
0.075x ( 0.075x (
0.050u 0.050u
Fl Fl
0.1 arcsec SVS13B 0.025 0.1 arcsec Per-emb-50 0.025
VLA 8mm Data Model Residual 0.200 VLA 8mm Data Model Residual 0.200
0.175 0.175
) )
0.150am 0.150am
0.125be 0.125be
0.100Jy/ 0.100Jy/
m m
0.075x ( 0.075x (
0.050u 0.050u
Fl Fl
0.1 arcsec Per-emb-14 0.025 0.1 arcsec Per-emb-30 0.025
VLA 8mm Data Model Residual 0.200
0.175
)
0.150am
0.125be
0.100Jy/
m
0.075x (
0.050u
Fl
0.1 arcsec HH211 mms 0.025
-
VLA 8mm Data Model Residual 0.200
0.175
)
0.150am
0.125be
0.100Jy/
m
0.075x (
0.050u
Fl
0.1 arcsec Per-emb-8 0.025
Figure 1. VLAA+Barraydata(left),q=0.25modelfromu,v-planebest-fit(center), andresidual(right). Images wereproducedwith
robust=0.25weighting. Contours startat3σ (σ 15µJy) withafactorof√2spacing. Thesynthesizedbeam isinthelowerleft.
∼
Table 1
Observations
Source α δ A-array B-array CombinedBeam BeamP.A.
(J2000) (J2000) Obs. Date Obs. Date (mas mas) (◦)
×
SVS13B 03:29:03.078 +31:15:51.740 03May2014 22Oct2013 105 83 -74.8
×
Per-emb-50 03:29:07.768 +31:21:57.125 01Jun2014 26Oct2013 97 94 54.0
×
Per-emb-14 03:29:13.548 +31:13:58.153 25Feb2014 22Oct2013 91 75 82.4
×
Per-emb-30 03:33:27.303 +31:07:10.161 16May2014 23Oct2013 99 92 -71.5
×
HH211-mms 03:43:56.805 +32:00:50.202 24Feb2014 21Oct2013 96 77 85.4
×
IC348MMS 03:43:57.064 +32:03:04.789 23Mar2014 22Oct2013 89 80 -57.7
×
Per-emb-8 03:44:43.982 +32:01:35.209 22Feb2014 01Oct2013 82 72 -71.6
×
Note. —Observationdates markthestartoftheobservations. Combinedbeamsizesreflectrobust=0.25weighting.
tively; Tobin et al. 2016, in press). The early results of the Shakura-Sunyaev disk model. VANDAM survey
the VANDAM survey have already revealedseveralcan- data provide an unparalleled opportunity to study the
didate disk sources in both Class 0 and I sources at youngest disks: the unsurpassed resolution, sample size,
10 AU scales. We consider these to be candidate disks and sensitivity for Class 0 and I protostars, permit a
because we lack kinematic data on small scales to de- detailed examination of typical continuum properties of
termine whether these structures are rotationally sup- young disks.
ported. Nevertheless, so little is known about this early
stage of disk evolution that even continuum-only imag-
ing of young disks provides useful constraints on their 2. OBSERVATIONS
properties.
The VANDAM survey includes Ka-band lower-
Inthis Letter,we presentthe firstresults towardsome
resolution (∼0.28′′ or 65 AU) B-array data and
of the protostellar disk candidates around Class 0 and
high-resolution (∼0.05′′ or 12 AU) A-array data. We
Class I sources from the VANDAM survey. We show
detected 16 protostellar candidate disk sources in the
the observed structures are consistent with disks by
Perseusmolecular cloud with the data collectedin 2013,
fitting the 8 mm dust-continuum data in the u,v-plane
2014, and 2015 (Segura-Cox et al. in prep). Here we
to a simple, parameterized emission model based on
focus on seven of the candidate disks (Table 1; Figure
VANDAM Class 0/I Candidate Disks 3
Table 2
SourceData
Source Class Deconvolved Size DiskP.A. DiskInclination F8mm Md F>1700kλ
(mas mas) (◦) (◦) (µJy) (M⊙) (µJy)
×
SVS13B 0 163 80 71.4 2.5 61 1352.7 11.6 0.14-0.29 82.0
× ± ±
Per-emb-50 I 137 53 170.0 0.3 67 1664.9 12.5 0.18-0.36 133.2
× ± ±
Per-emb-14 0 174 76 12.7 0.9 64 882.1 13.0 0.09-0.19 68.8
× ± ±
Per-emb-30 0 87 74 40.0 23.0 31 957.0 9.4 0.10-0.21 130.9
× ± ±
HH211-mms 0 93 59 34.8 9.6 51 867.5 8.1 0.09-0.19 42.9
× ± ±
IC348MMS 0 145 105 70.8 2.2 44 1126.5 10.3 0.12-0.24 0.0
× ± ±
Per-emb-8 0 111 84 116.1 2.8 41 1120.7 10.3 0.12-0.24 126.5
× ± ±
Note. — Sizes and angles are measured from image-plane 2D Gaussian fits. Angles are measured counterclockwise from north.
Uncertaintiesonthedeconvolved sizesare 5.0mas. Uncertaintiesoninclinationsare 10◦.
∼ ∼
1, left) which were observed with A-array prior to 2015 K, and we determined upper estimates using T = 20
d
andcanbe wellmodeledwith the prescriptiondescribed K. These candidate disks range in flux from 867.5 µJy
in Section 4. The observations were made in three-bit to 1664.9 µJy, providing estimated masses of 0.09–0.36
correlatormode,withabandwidthof8GHzdividedinto M⊙ (Table 2).
64 sub-bands. Each sub-band has 128 MHz bandwidth,
2MHzchannels,andfullpolarizationproducts. Thetwo
4 GHz basebands are centered at 36.9 GHz (∼8.1 mm) 4. MODELING
and29.0GHz(∼10.5mm). Threesourceswereobserved
We modeled the 8 mm A+B array continuum data
ineach3.5hour block,andtwosourceswereobservedin
by fitting a symmetric disk intensity profile to the con-
each2.75hourblock. Someobservationswereconducted
tinuum emission of each source. In this modeling, we
as 1.5 hour blocks. 3C48 served as the flux calibrator,
deprojected the 8 mm visibility data to a fixed position
and3C84wasthebandpasscalibrator. Theobservations
angle(P.A.)andfixedinclinationasdeterminedthrough
were taken in fast-switching mode to account for rapid
image-plane 2D Gaussian fitting of the disk candidates
atmospheric phase variations, with a 2.5 minute total
and assumed the disks are circularly symmetric (Table
cycle time to switch between the target source and
2). We azimuthally averaged the visibility data in the
the complex gain calibrator, J0336+3218. The total u,v-plane and binned the data in linearly spaced bins
integration time on each source was ∼30 minutes for
of width 50 kλ from 0 to 1500 kλ, switching to 30 log-
both A-array and B-array. We reduced the data with
spaced bins from 1500 to 4000 kλ where the data be-
CASA 4.1.0 and the VLA pipeline (version 1.2.2). We
comes noisier in order to boost the signal-to-noise level
executed additional flagging beyond pipeline flagging at large u,v-distances. We accounted for a lower-limit
by inspecting the phase, gain and bandpass calibration
free-freepointsourcecomponentemanating fromshocks
solutions. VLA Ka-band data sets have an estimated
in the protostellar jets (Anglada et al. 1998). Because
amplitude calibration uncertainty of ∼10%, but only
point sources in the image domain have constant flux
statistical uncertainties are considered in our analysis. density at all u,v-distances, we account for the free-free
componentbycalculating the averagerealcomponentof
the binned data with u,v-distance >1700kλ and include
3. ESTIMATEDMASSES theaverageasaflat,linearcomponentofthemodel(Ta-
Twokeyquantitiesthatdescribedisksaretheirmasses ble 2); the profiles become flat at values >1700 kλ in all
and radii. While accurately determining disk radii re- seven sources (e.g., Figure 2).
quiresmodeling,massesareestimatedfromfluxmeasure- We fitted the real components of the deprojected,
ments(Table2). Byassumingopticallythinemission,we averaged, and binned profile to a simple disk model
estimatediskmassesfromthe8mmdustcontinuumflux using a C-based implementation of emcee, an affine-
with the relation (Hildebrand 1983): invariant Markov chain Monte Carlo ensemble sampler
(Goodman & Weare2010;Foreman-Mackey et al.2013).
d2Fν The imaginary components were assumed to be zero in
M = , (1)
B(T )κ the model since the disk is positioned at the phase cen-
d ν
ter and assumed to be symmetric. The model mimics a
where Fν, d, κν, and Bν(Td), are respectively the total Shakura-Sunyaevdisk(Shakura & Sunyaev1973)witha
observed flux, distance, grain opacity, and blackbody power law temperature profile; the resultant model disk
intensity at dust temperature Td. We estimate κν at surface brightness profile is
8.1 mm by normalizing to Ossenkopf & Henning (1994)
(a1t/11.030)m(νm/23u1sGinHgza))βducstm2togg−a1s. ratβio=of11/i1s00ty:pκicνall=y I(r)∝ r −(γ+q)exp r (2−γ) , (2)
(cid:18)R (cid:19) (cid:26)(cid:18)R (cid:19) (cid:27)
c c
assumed for protostellar disks (Andrews et al. 2009),
giving κ =0.00146 cm2 g−1. Mass estimates are inher- where I(r) is the radial surface brightness distribution,
ν
ently ambiguous within an order of magnitude due to q is the temperature exponent, γ is the power-law of
uncertainties in the dust-to-gas ratio, T , and β; rather the inner-disk surface density, and r is radius. R is a
d c
than compute a single mass estimate, we calculate lower characteristic radius at which there is significant optical
and upper mass estimates for each source by varying depth and disk material, making it a proxy for outer
T . Lower masses were estimated by assuming T = 40 disk radius. This flux density profile is appropriate
d d
4 Segura-Cox et al.
angular scale (arcsec)
3.355 0.503 0.252 0.168 0.126 0.101 0.084 0.072 0.063
1.0
Per-emb-14: q = 0.25 R 1.5
−
0.8 R 2.0
−
point source
) 0.6
y
J
m
(
t 0.4
n
e
n
o
p 0.2
m
o
c
0.0
al
e
r
−0.2
−0.4
y) 0.4
J
m
0.2
(

al 0.0
u
d −0.2
si
e −0.4
r
0 500 1000 1500 2000 2500 3000 3500 4000
uv-distance (kλ)
Figure 2. Sample realvs u,v-distance plotof 8mm data. Top: realcomponent of data. The bluedashed lineindicates realcomponent
ofzero. Theredsolidlineisthebest-fitmodel. Bottom: residualofrealcomponent minusmodel.
here because the 8 mm data is expected to be in the ues near one and the nearly empty residual maps indi-
Rayleigh-Jeans tail of the dust emission. Flux is given cate the seven candidate disk sources were well-modeled
∞
by F = I(r) 2πr dr. Flux, disk radius, and gradient by our simple-shaped disk profile and are hence likely
0
were freRe parameters in the fitting. We fitted a model to be Class 0 and young Class I disks rather than in-
disk for values of q = [0.25,0.50,0.75,1.00] to the ner envelope structure (Chiang et al. 2008). All candi-
observationsin orderto avoidover-fittingthe data while date disks have modeled R > 10 AU, and except for
c
exploring reasonable physical values of q. HH211-mmsthe disks are largecomparedto the R = 10
AUupperlimitpredictedbymagneticbrakingmodelsof
Dapp & Basu (2010) at the Class 0 stage.
5. RESULTS Themodelincludesadiskandpoint-sourcecomponent
but does not take into account envelope emission. The
Once model fitting was completed on the candidate
observationsfilter outthe majorityofthe contaminating
disks, we generateddisk model visibilities corresponding
envelope emission, demonstrated by the visibility pro-
to the best-fit parameters. The best-fit model andresid-
files corresponding to R−1.5 and R−2.0 envelope surface
uals for each source were Fourier transformed, sampled
density profiles plotted in Figure 2. These envelope
at the same u,v-points as the data, and imaged using
density profiles represent, respectively, free-fall collapse
the same weighting as the data (Figure 1, center). We
and the singular isothermal sphere (Shu 1977). The
subtractedthe modelvisibility fromthe datato produce
best-fit models were determined from the maximum
a residual visibility set. Synthetic maps were created
likelihood, i.e. the lowest χ2 value, of all models fit to
of the best-fit model disks. Residual maps were made
the data. Values of q near 0.5 are predicted by theory
byimagingtheFouriertransformedvisibilitiesaftersub-
(e.g., Chiang & Goldreich 1997) for Class II protostars,
tractingthemodelfromthedataintheu,v-plane(Figure
which are more evolved than the protostars in our
1, right). An example of the binned observational data,
sample. Lower values of q result in lower values of
model fits, and residuals is also shown in Figure 2.
Fit results are shown in Table 3. The χ2 val- χ2reduced for six candidate disks. Values of q < 0.5 for
reduced
VANDAM Class 0/I Candidate Disks 5
valuesfoundforolderClassIIsources. Thebest-fitmod-
Table 3
8mmBest-fitModelingResults els for our seven sources yield −0.58 < γ < 0.45. Nega-
Source q γ Rc χ2reduced triavdeiuvsa,liunecsoonfsiγsteinndtiwcaitthedinisckrse,aysientgfosruraflalcseoudrecnessitaytlweaitsht
(AU)
SVS13B 0.25 0.21+0.23 24.28+2.1 2.194 one value of q exists whichproduces a positive best-fit γ
−0.20 −1.7
0.50 0.42+0.25 25.50+1.9 2.185 (Table3). Andrews et al.(2010)determinedγ at880µm
−0.21 −1.5
0.75 0.63+0.24 26.46+1.6 2.175 fordisksaroundClassIIobjectsusingatwo-dimensional
−0.22 −1.4
1.00 0.85+0.26 27.28+1.4 2.164 parametricmodelwith a prescriptionofthe surfaceden-
Per-emb-50 0.25 0.08−+00..0223 21.9+−01..82 1.556 sityprofile similarto the radialsurfacebrightnessdistri-
0.50 0.26−+00..1156 23.3−+10..19 1.558 bution applied in this Letter. The Class II sources have
0.75 0.44−+00..1167 24.6−+11..40 1.560 values of 0.4 < γ < 1.1, larger than VANDAM results.
1.00 0.64−+00..1167 25.7−+11..41 1.563 The modeled, surface density profiles of the Class 0 and
Per-emb-14 0.25 -0.11−+00..1186 28.5+−21..33 1.110 I systems taper off less quickly with radius compared to
0.50 0.09+−00..0080 30.6−+22..81 1.114 the Class II systems inside the characteristic radius Rc.
−0.21 −2.3 The best-fit model radiiof the candidate Class 0 disks
0.75 0.27+−00..1274 32.5+−22..28 1.119 are between ∼15-30 AU, except HH211-mms with R
1.00 0.48+−00..1293 33.9+−33..61 1.123 ∼ 10 AU. These are smaller than the Keplerian diskcs
Per-emb-30 0.25 0.02+0.18 14.0+1.0 1.100
−0.31 −0.9 found in L1527 and VLA 1623 at 1.3 mm (Ohashi et al.
0.50 0.20+−00..0342 14.9+−11..91 1.102 2014; Murillo et al. 2013, R ∼ 54 AU and R ∼ 189
0.75 0.39+0.30 15.8+1.9 1.104
−0.34 −1.3 AU respectively) but consistent with the size of HH212
1.00 0.59+−00..1343 16.5−+311.6.3 1.107 (Codella et al.2014,R>30AU).Themodeledradii(Ta-
HH211-mms 0.25 0.48+−00..4708 10.5+−00..88 1.009 ble3)areafactorof1to1.5timeslargerthanthedecon-
0.50 0.65+−00..4832 11.0+−10..09 1.009 volved sizes (Table 2). The radii of the new candidate
0.75 0.81+−00..4729 11.5+−11..22 1.009 disks may represent lower limits on disk sizes because
1.00 1.01+−00..4841 11.9+−11..43 1.009 continuum emission from dust is biased by dust grain
IC348MMS 0.25 -0.58+−00..1111 25.7+−22..82 1.085 size. Radial drift (Birnstiel et al. 2010; Weidenschilling
0.50 -0.39+−00..1191 29.0+−32..26 1.096 1977) sends large grains inward in the disk, and so the
0.75 -0.19+−00..1217 31.6+−42..19 1.107 long wavelength observations preferentially trace inner
1.00 0.02+−00..0171 33.7+−43..31 1.118 disk emission (P´erez et al. 2012). We expect the 8 mm
Per-emb-8 0.25 0.01+−00..1169 19.0+−11..21 1.099 dust-modeled radii to be less than the gas disk radius,
0.50 0.20+−00..1270 20.2+−11..43 1.107 andsothe8mmradiiarelower-limitsondisksize;multi-
0.75 0.40+0.17 21.2+1.6 1.114 wavelength observations that include gas tracers are re-
−0.21 −1.4
1.00 0.61+0.17 22.1+1.8 1.122 quired to further constrain disk radii. Per-emb-14 was
−0.20 −1.6
resolvedat1.3mmbyTobin et al.(2015b)withR∼100
Note. — Values of q are fixed. Values of γ and Rc are deter- AU; thus the disk is indeed likely larger than the 8 mm
mined from best-fit models. Uncertainties reflect 90% confidence radius quoted here. In all cases, the previously known
intervals. disks and new candidate Class 0 disks are larger than
Class 0 and younger Class I sources are favored since the expected upper limit of10AU fromstrongmagnetic
the envelope mass reservoir is still large, and radiation
braking models (e.g., Dapp & Basu 2010).
is reprocessed from the protostar and directed back The seven VANDAM candidate disks are well-fit by a
onto the disk, increasingthe brightness ofthe outer disk disk-shaped model and have disk masses, values of γ,
relative to the inner regions of the disk and flattening
and radii consistent with known disks. These candi-
the brightnessdistribution(D’Alessio 1996). Compared date disks can be compared to the well-studied Class
to the other modeled protostars, χ2 is notably 0 systems with disks: L1527 (Tobin et al. 2012) and
reduced
higher for SVS13B which may be a more complicated VLA 1623 (Murillo & Lai 2013). L1527 and VLA 1623
source and not well described by the model. have large Keplerian disks (R ∼ 50 AU, Ohashi et al.
2014; Murillo et al. 2013) as well as misaligned mag-
netic fields and rotationaxes (Hull et al. 2014), suggest-
6. DISCUSSION
ing that misaligned fields and large disks may be linked
The estimated masses of the candidate disks are con- (Segura-Cox et al. 2015). The discrepancy between the-
sistent with known disks around Class 0 and I proto- oretical and observed disk sizes could be due to mis-
stars. ThesevenVANDAMcandidateClass0andIdisks alignment between the magnetic field and rotation axis,
range in estimated masses of 0.09–0.36 M⊙. Scaling to whichchangesthestrengthofmagneticbraking,allowing
our value of κν, Class 0 protostar L1527’s disk is 0.013 disks to grow at early times (Hennebelle & Ciardi 2009;
M⊙(Tobin et al.2013)withTd=30K.Arecentinterfer- Joos et al. 2012; Li et al. 2013; Krumholz et al. 2013).
ometric study of Class I disks in Taurus (Harsono et al. Similarly, aligned orientations would strengthen mag-
2014)revealedthe disksaroundTMC1A,TMC1,TMR1 neticbrakingandlimitdisksize,asseeninClass0source
and L1536 to have masses of 0.4–3.3×10−2 M⊙ with Td B335withadiskofR<5AU(Yen et al.2015;Hull et al.
= 30 K and Ossenkopf & Henning (1994) opacities. All 2014). The large Keplerian Class 0 disks and new can-
mass estimates have uncertainties of a factor of ∼10due didate disks indicate significant magnetic braking either
to the unknown gas-to-dust ratio, T , β, and varying hasalreadyoccurred,hasnothappened,orthemagnetic
d
choices of κ . field is weak enough for disks with R > 10 AU around
ν
Values of γ, the power-law of the inner-disk surface these Class 0 protostars to form, though other configu-
density, for the VANDAM sources can be compared to rations are possible.
6 Segura-Cox et al.
It is illustrative to compare the presence of VANDAM DAM candidate disk results combined with polarization
candidate disks to the magnetic field morphology of the data for two of the sources indicate that indeed R > 10
systems. The TADPOL survey examines protostellar AU disks have formed in systems with misaligned orien-
dust polarization and hence plane-of-sky magnetic field tations.
orientation. (Hull et al.2014). Forsourceswithdetected
magnetic fields, we are able to determine which systems
This research made use of APLpy, an open-
have misaligned orientations between the outflow (a
source plotting package for Python hosted at
proxy for rotation axis) and the average magnetic field
http://aplpy.github.com.
orientation. The magnetic field morphologies of two of
DMSC is currently supported by NRAO Student Ob-
the seven VANDAM sources in this Letter (HH211-mms
serving Support grant SBC NRAO 2015-06997. JJT
and SVS13B) were also examined as a part of the
is currently supported by grant 639.041.439 from
TADPOL survey. A qualitative comparison between
the Netherlands Organisation for Scientific Research
the VANDAM data and the TADPOL magnetic field
(NWO).ZYLissupportedbyNASANNX14AB38Gand
orientation reveals that HH211-mms and SVS13B mor-
NSF AST-1313083.
phologies agree with the tentative link between R > 10
The National Radio Astronomy Observatory is a
AU disks and misaligned orientations(Segura-Cox et al.
facility of the National Science Foundation operated
2015). HH211-mms has a clearly misaligned orientation
undercooperativeagreementbyAssociatedUniversities,
(Hull et al. 2014), agreeing with theoretical predictions
Inc.
that disks with R > 10 AU disks form at early times
in systems with non-parallel orientations. SVS13B has
a more complicated magnetic field morphology with an
average aligned orientation (Hull et al. 2014) with the
outflow (Bachiller et al. 1998), however the TADPOL REFERENCES
data has lower resolution than our data (∼3′′, or 700
AU), and the innermost regions of the magnetic field Andr´e,P.,Ward-Thompson,D.,&Barsony,M.2000,in
has misaligned orientation near the candidate disk. The ProtostarsandPlanets IV,59
small-scale, misaligned orientation in SVS13B could Andrews,S.M.,Wilner,D.J.,Hughes,A.M.,etal.2009,ApJ,
700,1502
be caused by magnetic fields being wrapped up with
Andrews,S.M.,Wilner,D.J.,Hughes,A.M.,etal.2010,ApJ,
the rotating disk, and the aligned magnetic field lines
723,1241
further from the disk are less influenced by the disk Anglada,G.,Villuendas,E.,Estalella,R.,etal.1998, AJ,116,
wrapping. 2953
Armitage,P.J.,2011,ARA&A,49,195
Bachiller,R.,Guilloteau,S.,Gueth, F.,etal.1998,A&A,339,49
7. CONCLUSIONS Birnstiel,T.,Ricci,L.,Trotta,F.,etal.2010,A&A,516,14
Chiang,E.I.&Goldreich,P.1997,ApJ,490,368
The firstresults of the VANDAM survey revealednew Chiang,H.-F.,Looney, L.W.,Tassis,K.,etal.2008,ApJ,680,
candidate Class 0 and I disks in continuum emission in 474
the Perseus molecular cloud and is the most sensitive, Codella,C.,Cabrit,S.,Gueth,F.,etal.2014,A&A,568,5
D’Alessio,P.1996,PhDthesis,Univ.NacionalAst´onomade
most complete, and highest resolution survey of young
M´exico
protostellar disks to-date. For seven of the VANDAM Dapp,W.B.,&Basu,S.2010,A&A,521,L56
systems, we fit the deprojected, averaged, and binned Dapp,W.B.,Basu,S.,&Kunz,M.W.,2012,A&A,541,A35
data in the u,v-plane to a disk-shaped profile, account- Dunham,M.M.,Stutz, A.M.,Allen,L.E.,etal.2014,
ing for a free-free point source potentially contaminat- ProtostarsandPlanets VI,195
Foreman-Mackey,D.,Hogg,D.W.,Lang,D.,Goodman, J.2013
ing the dust emission, to affirm the disk candidacy of
PASP,125,306
the sources and to begin to model disk properties. To Goodman,J.&Weare,J.,2010,Comm.App.Math.Comp.Sci.,
confirmthat these disks are true,rotationally-supported 5,65
disks,kinematicfollow-upobservationsarerequired. The Harsono,D.,Jørgensen,J.K.,vanDishoeck,E.F.,etal.2014,
seven VANDAM candidate disks are well-fit by a model A&A,562,77
Hennebelle,P.,&Ciardi,A.2009, A&A,506,29
with a disk-shaped profile and have masses comparable
Hildebrand,R.H.1983,QJRAS,24,267
to known disks. The inner-disk surface densities of the Hirota,T.,Bushimata,T.,Choi,Y.K.,etal.2008,PASJ,60,37
Class 0 and I candidate disks taper off less quickly with Hirota,T.,Honma,M.,Imai,H.,etal.2011, PASJ,63,1
radius than their Class II counterparts. Hull,C.L.H.,Plambeck,R.L.,Kwon,W.,etal.2014,ApJS,
The best-fit model radii of the new candidate disks 213,13
are R > 15 AU, except HH211-mms with R ∼ 10 AU. Joos,M.,Hennebelle,P.,&Ciardi,A.2012, A&A,543,A128
c c Joos,M.,Hennebelle,P.,Ciardi,A.&Fromang,S.2013, A&A,
Since magnetic braking is expected to suppress Class 0 554,A17
diskstoR<10AU,thissuggeststhatmagneticbraking Jørgensen,J.K.,vanDishoeck,E.F.,Visser,R.,etal.2009,
has already ended or is inconsequential in some Class 0 A&A,507,861
Krumholz,M.R.,Crutcher,R.M.,&Hull,C.L.H.2013,ApJ,
sources. At 8 mm, the modeled radii are lower limits
767,L11
on the disk size since dust continuum emission is tied
Li,Z.-Y.,Krasnopolsky,R.,&Shang,H.2011,ApJ,738,180
to grain size and large grains radially drift inwards in Li,Z.-Y.,Krasnopolsky,R.,&Shang,H.2013,ApJ,774,82
the disk; the actual sizes of the disks may be larger. Looney,L.W.,Mundy,L.G.,&Welch,W.J.2000,ApJ,529,477
Multi-wavelength observations are needed to put added Machida,M.N.,Inutsuka, S.-I.,&Matsumoto,T.2011,PASJ,
63,555
constraints on disk radii. Furthermore, theory predicts
Machida,M.N.,Inutsuka, S.-I.,&Matsumoto,T.2014,MNRAS,
larger disk growth in systems with misaligned outflow
438,2278
axesandmagneticfieldorientations,andtheearlyVAN- Maury,A.J.,Andr´e,P.,Hennebelle,P.,etal.2010,A&A,512,40
VANDAM Class 0/I Candidate Disks 7
Mellon,R.R.,&Li,Z.-Y.2008,ApJ,681,1356 Tobin,J.J.,Hartmann,L.,Chiang,H.-F.,etal.2012,Nature,
Murillo,N.M.,&Lai,S.-P.2013,ApJL,764,L15 492,83
Murillo,N.M.,Lai,S.-P.,Bruderer,S.,etal.2013,A&A,560,103 Tobin,J.J.,Looney, L.W.,Li,Z.-Y.,etal.2016, ApJ,inpress
Ohashi,N.,Saigo,K.,Aso,Y.,etal.2014,ApJ,796,131 Tobin,J.J.,Dunham,M.M.,Looney, L.W,etal.2015a, ApJ,
Ossenkopf,V.,&Henning,T.1994, A&A,291,943 798,61
P´erez,L.M.,Carpenter,J.M.,Chandler,C.J.,eta.,2012, Tobin,J.J.,Looney, L.W.,Wilner,D.J.,etal.2015b,ApJ,805,
ApJL,760,L17 125
Segura-Cox,D.M.,Looney, L.W.,Stephens, I.W.,etal.2015, Weidenschilling,S.J.1977,MNRAS,180,57
ApJ,798,L2 Williams,J.P.&CiezaL.A.,2011, ARA&A,49,67
Seifried,D.,Banerjee,R.,Pudritz,R.E.,&Klessen,R.S.2013, Yen,H.-W.,Koch,P.M.,Takakuwa,S.,etal.2015,ApJ,799,193
MNRAS,432,3320
Shakura,N.I.,Sunyaev, R.A.1973,A&A,24,337
Shu,F.H.1977,ApJ,214,488
Tobin,J.J.,Bergin,E.A.,Hartmann,L.,etal.2013, ApJ,765,
18