A HUNT FOR MASSIVE STARLESS CORES

, , , , , and

Published 2017 January 13 © 2017. The American Astronomical Society. All rights reserved.
, , Citation Shuo Kong et al 2017 ApJ 834 193 DOI 10.3847/1538-4357/834/2/193

Download Article PDF
DownloadArticle ePub

You need an eReader or compatible software to experience the benefits of the ePub3 file format.

0004-637X/834/2/193

ABSTRACT

We carry out an ALMA ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) and 1.3 mm continuum survey of 32 high-mass surface density regions of seven infrared dark clouds, with the aim of finding massive starless cores that may form the initial conditions for the formation of massive stars. Cores showing strong ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) emission are expected to be highly deuterated and indicative of early, potentially pre-stellar stages of star formation. We also present maps of these regions in ancillary line tracers, including C18O(2-1), DCN(3-2), and DCO+(3-2). Over 100 ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ cores are identified with our newly developed core-finding algorithm, based on connected structures in position–velocity space. The most massive core has $\sim 70\ {M}_{\odot }$ (potentially $\sim 170\ {M}_{\odot }$) and so may be representative of the initial conditions or early stages of massive star formation. The existence and dynamical properties of such cores constrain massive star formation theories. We measure the line widths and thus velocity dispersion of six of the cores with strongest ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line emission, finding results that are generally consistent with virial equilibrium of pressure confined cores.

Export citation and abstract BibTeX RIS

1. INTRODUCTION

Massive star formation remains an important unsolved problem in astrophysics. Here, we seek to obtain improved observational constraints on the initial conditions and early stages of the process. If there is a universal star formation mechanism, such that massive stars ($\gt 8\,{M}_{\odot }$) are born via a scaled-up version of the low-mass Core Accretion mechanism (e.g., McLaughlin & Pudritz 1997; McKee & Tan 2003, hereafter MT03), then the initial conditions (i.e., at the time just before protostar formation) should be massive starless cores. Here, the term "core" is defined to be the self-gravitating structure that will collapse to a single, central, rotationally supported disk that eventually forms a single star or small N multiple. Early stages of massive star formation by this mechanism would include a low-mass protostar undergoing relatively ordered accretion fed by quasi-monolithic collapse, near the center of a massive core. The existence of such cores is a key difference between this model and competitive accretion (e.g., Bonnell et al. 2001; Wang et al. 2010), which involves fragmentation of gas into protostellar seeds with initial masses only on the order of the thermal Jeans mass—typically much less than a solar mass in the high-mass surface density, high-pressure clumps where massive stars form. Note that the term "clump" is defined to mean the self-gravitating cloud that eventually fragments into a star cluster. Only later do some of these seeds accumulate further material, fed from the collapsing clump, to become massive stars. Thus, finding and characterizing massive starless and early-stage cores is a key way to distinguish between massive star formation theories.

However, because massive stars are rare, massive starless/early-stage cores—even if they exist, they would also be rare, and thus, typically distant and relatively small in angular size. Furthermore, they would likely be surrounded by much larger quantities of cold, dense molecular clump gas, with most mass going into lower mass stars or being dispersed back into the diffuse interstellar medium. Finding massive starless/early-stage cores is, thus, a challenging problem.

We have developed a strategy to overcome this challenge. We target regions based on mid-infrared extinction mapping of infrared dark clouds (IRDCs) (Butler & Tan 2009, 2012 hereafter BT09, BT12), which probes mass surface densities up to ${\rm{\Sigma }}\,\sim \,0.5\ {\rm{g}}\,{\mathrm{cm}}^{-2}$, with an angular resolution of 2''. This allows detailed study of the structure of dense clumps: BT12 characterized 42 high Σ clumps selected from 10 IRDCs (A-J), which were chosen to be relatively nearby and dense. The 42 clumps were checked to make sure they are free of 8 and $24\ \mu {\rm{m}}$ (Spitzer-IRAC & MIPS) sources. We note that this method of sample selection differs from that based on following up strong mm continuum sources and then selecting those that are IR, including 70 μm, dark (e.g., Tackenberg et al. 2012; Traficante et al. 2015; Svoboda et al. 2016).

Our goal, which we carry out in this paper, is to search the majority of these sources for ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line emission. The abundance of this species is known to increase in cold, dense conditions of low-mass starless cores (e.g., Crapsi et al. 2005; Bergin & Tafalla 2007; Caselli & Ceccarelli 2012), where CO is largely frozen-out onto dust grain ice mantles and thus depleted from the gas phase. The enhanced abundance of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ with respect to ${{\rm{N}}}_{2}{{\rm{H}}}^{+}\,$is relatively well-understood from the astrochemical point of view, and we have developed a comprehensive spin-state, gas phase reaction network to model this deuteration process (Kong et al. 2015, hereafter K15). It is this high abundance of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ that acts as a signpost for the presence of a starless core on the verge of collapse, or an early-stage core just after protostar formation, allowing us to find these relatively rare locations in IRDCs.

We tested this method by observing four target regions centered on IRDC clumps with ALMA in Cycle 0, detecting 6 ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores at 2'' resolution (≥1 from each region) (Tan et al. 2013, hereafter T13). The two most massive cores were found in the IRDC clump C1: C1-N and C1-S. We estimated the masses of the cores, defined by projection of their 3σ $l-b-v$ space ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) contour, in two ways: (1) from the MIREX map, finding C1-N has $61\pm 30\ {M}_{\odot }$ and C1-S has $59\pm 30\ {M}_{\odot }$ with the ∼50% systematic uncertainty due to assumed distance (5 ± 1 kpc) and dust opacity (∼30%) uncertainties; (2) from mm dust continuum emission, finding C1-N has ${16}_{7}^{33}\ {M}_{\odot }$ and C1-S has ${63}_{27}^{129}\ {M}_{\odot }$, with uncertainties mostly due to the adopted dust temperature of $T=10\pm 3$ K, along with distance and dust emissivity uncertainties. Note that it is possible that, in general, the "core" may extend beyond the observed ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) contour, so these may be lower limits on the core mass. On the other hand, it is also possible that the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) structure may actually contain more than one core, i.e., it may be resolved into two or more separate cores, if observed at higher angular resolution.

Thus, of the six T13 cores, C1-S and C1-N are the most promising examples of a massive starless/early-stage cores, i.e., with $\gtrsim 20\ {M}_{\odot }$ that may allow formation of a $\gtrsim 10\ {M}_{\odot }$ star, given expected outflow regulated formation efficiencies ∼50% (Zhang et al. 2014). C1-S appears monolithic, centrally-concentrated in both ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) and mm continuum emission, and rounded (most likely by self-gravity). C1-N appears to be less centrally concentrated and potentially fragmented. Follow-up observations of other ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ and ${{\rm{N}}}_{2}{{\rm{H}}}^{+}\,$lines allowed measurement of ${D}_{\mathrm{frac}}\equiv [{{\rm{N}}}_{2}{{\rm{D}}}^{+}]/[{{\rm{N}}}_{2}{{\rm{H}}}^{+}]$ in the cores, with values of 0.2–0.7 (Kong et al. 2016). For most chemodynamical models, such high values (orders of magnitude greater than the cosmic [D]/[H] ratio of $\sim {10}^{-5}$) imply relatively old astrochemical ages, and thus, relatively slow collapse rates, $\lesssim 1/3$ of the rate of free-fall collapse.

Further follow-up with ALMA in Cycle 2 of the C1 region found the presence of a very collimated protostellar outflow, traced by 12CO(2-1), from a source within C1-S (in both position and velocity space), such that this is most likely to be an example of an early-stage massive core (Tan et al. 2016). A second protostellar outflow source also overlaps spatially with C1-S, although its association with the core in velocity space is less certain. No outflows were seen from C1-N.

T13 used the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line-width to study the dynamics of the cores. For the sample of six sources, the velocity dispersions were, on average, consistent (within a factor of ∼0.8) with those expected from virial equilibrium of the fiducial MT03 Turbulent Core model. However, for C1-S, the observed velocity dispersion is about a factor of two smaller than the fiducial virial equilibrium prediction. If virial equilibrium is being maintained, as would be expected if the astrochemical age is larger than the dynamical time, then relatively strong magnetic fields, ∼1 mG, are needed.

We see that larger samples of starless and early-stage cores are needed to better test the different theoretical models. This has motivated the observations and analysis presented in this paper.

2. SAMPLE SELECTION AND ALMA OBSERVATIONS

BT09 and BT12 studied the 10 IRDCs A-J, selecting them from the sample of Rathborne et al. (2006) to be particularly suitable for MIREX mapping: they are relatively nearby (thus reducing the contribution of MIR foreground emission), show high contrast against the Galactic MIR background, and are surrounded by relatively simple, smooth MIR background emission. BT12 analyzed the detailed structural properties of 42 clumps selected from these clouds to be dark at 8 and 24 μm, the latter evaluated from the MIPSGAL survey (Carey et al. 2009). T13 observed ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) in 4 clumps, C1, F1, F2, G2, with ALMA in Cycle 0.

With ALMA in Cycle 2 (Project number: 2013.1.00806.S; PI: Tan), we observed 32 more clumps from the BT12 sample (listed in Table 1), focusing on IRDCs A, B, C, D, E, F, and H. The observations were carried out during 2015 April, with the 12 m array in the most compact configuration. The baselines were from 12 to 330 m (9–254 kλ), resulting in a synthesized beam size of 1farcs× 1farcs0 and a maximum detectable scale of ∼20''. The diameter of the primary beam, which approximately sets the field-of-view (FOV), is ∼26''. The maximum scale is comparable to the FOV, and ALMA has very good uv-coverage in the short spacings in the compact configuration. No ACA observations were performed. Two of the target pointings contained two BT12 clumps; in total, there were 30 pointings in our observations (two tracks, each of 15 pointings). Together with the four targets already observed, this completes 86% of the BT12 sample (92% of the sources in IRDCs A-H, which are at a range Galactic longitudes from l = 18fdg8 to 35fdg5).

Table 1.  IRDC Clump Targets of the ALMA Cycle 2 Observation

Clumpa R.A. Decl. l (deg) b (deg) vLSR ($\mathrm{km}\ {{\rm{s}}}^{-1}$) d (kpc)
A1 ${18}^{{\rm{h}}}{26}^{{\rm{m}}}15\buildrel{\rm{s}}\over{.} 14$ −12°41'43farcs4 18.78675 −0.28592 66 4.8
A2 ${18}^{{\rm{h}}}{26}^{{\rm{m}}}19\buildrel{\rm{s}}\over{.} 04$ −12°41'14farcs8 18.80117 −0.29625 66 4.8
A3 ${18}^{{\rm{h}}}{26}^{{\rm{m}}}21\buildrel{\rm{s}}\over{.} 78$ −12°41'10farcs2 18.80750 −0.30550 66 4.8
B1 ${18}^{{\rm{h}}}{25}^{{\rm{m}}}52\buildrel{\rm{s}}\over{.} 83$ −12°04'52farcs7 19.28758 0.08083 26 2.4
B2 ${18}^{{\rm{h}}}{25}^{{\rm{m}}}58\buildrel{\rm{s}}\over{.} 29$ −12°04'13farcs5 19.30758 0.06625 26 2.4
C2 ${18}^{{\rm{h}}}{42}^{{\rm{m}}}50\buildrel{\rm{s}}\over{.} 45$ −04°03'17farcs7 28.34383 0.06017 79 5.0
C3 ${18}^{{\rm{h}}}{42}^{{\rm{m}}}44\buildrel{\rm{s}}\over{.} 02$ −04°01'54farcs5 28.35217 0.09450 79 5.0
C4 ${18}^{{\rm{h}}}{42}^{{\rm{m}}}49\buildrel{\rm{s}}\over{.} 34$ −04°02'27farcs3 28.35417 0.07067 79 5.0
C5 ${18}^{{\rm{h}}}{42}^{{\rm{m}}}52\buildrel{\rm{s}}\over{.} 59$ −04°02'44farcs3 28.35617 0.05650 79 5.0
C6 ${18}^{{\rm{h}}}{42}^{{\rm{m}}}54\buildrel{\rm{s}}\over{.} 37$ −04°02'31farcs7 28.36267 0.05150 79 5.0
C7 ${18}^{{\rm{h}}}{42}^{{\rm{m}}}40\buildrel{\rm{s}}\over{.} 01$ −04°00'34farcs4 28.36433 0.11950 79 5.0
C8 ${18}^{{\rm{h}}}{42}^{{\rm{m}}}59\buildrel{\rm{s}}\over{.} 99$ −04°01'33farcs1 28.38783 0.03817 79 5.0
C9 ${18}^{{\rm{h}}}{42}^{{\rm{m}}}51\buildrel{\rm{s}}\over{.} 86$ −03°59'43farcs3 28.39950 0.08217 79 5.0
D1 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}17\buildrel{\rm{s}}\over{.} 05$ −04°02'01farcs5 28.52717 −0.25033 87 5.7
D2 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}23\buildrel{\rm{s}}\over{.} 79$ −04°02'11farcs5 28.53750 −0.27650 87 5.7
D3 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}15\buildrel{\rm{s}}\over{.} 38$ −04°00'50farcs7 28.54150 −0.23517 87 5.7
D4 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}22\buildrel{\rm{s}}\over{.} 51$ −04°01'53farcs5 28.53950 −0.26950 87 5.7
D5 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}16\buildrel{\rm{s}}\over{.} 43$ −03°59'22farcs3 28.56533 −0.22783 87 5.7
D6 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}17\buildrel{\rm{s}}\over{.} 78$ −04°00'12farcs4 28.55550 −0.23917 87 5.7
D7 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}17\buildrel{\rm{s}}\over{.} 69$ −03°59'26farcs5 28.56667 −0.23300 87 5.7
D8 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}18\buildrel{\rm{s}}\over{.} 29$ −03°59'06farcs2 28.57283 −0.23267 87 5.7
D9 ${18}^{{\rm{h}}}{44}^{{\rm{m}}}18\buildrel{\rm{s}}\over{.} 74$ −03°58'13farcs7 28.58667 −0.22767 87 5.7
E1 ${18}^{{\rm{h}}}{43}^{{\rm{m}}}06\buildrel{\rm{s}}\over{.} 71$ −03°45'09farcs9 28.64350 0.13817 80 5.1
E2 ${18}^{{\rm{h}}}{43}^{{\rm{m}}}10\buildrel{\rm{s}}\over{.} 12$ −03°45'15farcs8 28.64850 0.12483 80 5.1
F3 ${18}^{{\rm{h}}}{53}^{{\rm{m}}}18\buildrel{\rm{s}}\over{.} 42$ +01°27'33farcs9 34.44383 0.24967 58 3.7
F4 ${18}^{{\rm{h}}}{53}^{{\rm{m}}}18\buildrel{\rm{s}}\over{.} 51$ +01°28'30farcs5 34.45800 0.25650 58 3.7
H1 ${18}^{{\rm{h}}}{57}^{{\rm{m}}}11\buildrel{\rm{s}}\over{.} 37$ +02°07'27farcs1 35.47800 −0.31033 44 2.9
H2 ${18}^{{\rm{h}}}{57}^{{\rm{m}}}06\buildrel{\rm{s}}\over{.} 92$ +02°08'20farcs9 35.48283 −0.28700 44 2.9
H3 ${18}^{{\rm{h}}}{57}^{{\rm{m}}}08\buildrel{\rm{s}}\over{.} 83$ +02°08'24farcs3 35.48733 −0.29367 44 2.9
H4 ${18}^{{\rm{h}}}{57}^{{\rm{m}}}06\buildrel{\rm{s}}\over{.} 88$ +02°08'44farcs8 35.48867 −0.28383 44 2.9
H5 ${18}^{{\rm{h}}}{57}^{{\rm{m}}}08\buildrel{\rm{s}}\over{.} 27$ +02°08'57farcs7 35.49450 −0.28733 44 2.9
H6 ${18}^{{\rm{h}}}{57}^{{\rm{m}}}08\buildrel{\rm{s}}\over{.} 17$ +02°10'51farcs8 35.52250 −0.27250 44 2.9

Note.

aTargets selected from BT12.

Download table as:  ASCIITypeset image

Dual polarization mode was adopted for the Band 6 spectral setup. Four basebands and seven spectral windows were used during the observations. A Baseband 1 single spectral window was centered on ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) (rest frequency 231.32 GHz), with a velocity resolution of 0.05 $\mathrm{km}\ {{\rm{s}}}^{-1}$. A Baseband 2 single spectral window was used for a continuum observation, centered at 231.00 GHz. The total bandwidth for this baseband is about 2 GHz. A Baseband 3 single spectral window was centered on ${{\rm{C}}}^{18}{\rm{O}}$(2-1) (rest frequency 219.56 GHz), with a velocity resolution of 0.05 $\mathrm{km}\ {{\rm{s}}}^{-1}$. Baseband 4 was split into four spectral windows, including a window at 216.11 GHz for DCO+(3-2), a window at 216.95 GHz for CH3OH(v t = 0 5(1, 4)-4(2, 2), with upper-state energy of 56 K), a window at 217.10 GHz for SiO(5-4), and a window at 217.24 GHz for DCN(3-2). Each of these lines has a 0.2 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$velocity resolution.

This paper will focus mostly on the results from the continuum and ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) observations, although the integrated intensity maps of most of the other species are also presented. The SiO(5-4) data, which probe protostellar outflows, are presented by M. Liu et al. (2016, in preparation).

The sample of 30 targets was divided into two tracks, each containing 15 sources. Track 1, with reference velocity of +58 $\mathrm{km}\ {{\rm{s}}}^{-1}$, includes A1, A2, A3, (${v}_{\mathrm{LSR}}\simeq +66\ \mathrm{km}\ {{\rm{s}}}^{-1}$: these estimates are derived from 13CO(1-0) emission from the clouds: see, e.g., Hernandez & Tan 2015), B1, B2 (${v}_{\mathrm{LSR}}\simeq +26\ \mathrm{km}\ {{\rm{s}}}^{-1}$), C2,7 C3, C4, C5, C6, C7, C8, C9 (${v}_{\mathrm{LSR}}\simeq +79\ \mathrm{km}\ {{\rm{s}}}^{-1}$), E1, and E2 (${v}_{\mathrm{LSR}}\simeq +80\ \mathrm{km}\ {{\rm{s}}}^{-1}$). Track 2, with reference velocity of +66 $\mathrm{km}\ {{\rm{s}}}^{-1}$, includes D1, D2 (also contains D4), D3, D5 (also contains D7), D6, D8, D9 (${v}_{\mathrm{LSR}}\simeq +87\ \mathrm{km}\ {{\rm{s}}}^{-1}$), F3, F4 (${v}_{\mathrm{LSR}}\simeq +58\ \mathrm{km}\ {{\rm{s}}}^{-1}$), H1, H2, H3, H4, H5, and H68 (${v}_{\mathrm{LSR}}\simeq +44\ \mathrm{km}\ {{\rm{s}}}^{-1}$). In Track 1, J1924-2914 was used as the bandpass calibrator, J1832-1035 was used as the gain calibrator, and Neptune was the flux calibrator. In Track 2, J1751+0939 was used as the bandpass calibrator, J1851+0035 was used as  the gain calibrator, and Titan was the flux calibrator. The continuum data were cleaned, but the line data were not cleaned due to their relatively weak detections. No self-calibration was done to the data set. Primary beam correction was applied before fluxes were extracted.

Our sensitivity level was determined by the desire to detect massive ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ cores that are similar to C1-S with $\gtrsim 5\sigma $ significance integrating over the typical velocity range of such cores, ∼1 $\mathrm{km}\ {{\rm{s}}}^{-1}$. We estimated this to be a sensitivity level of 30 mJy per beam per 0.1 $\mathrm{km}\ {{\rm{s}}}^{-1}$, i.e., three times worse than that achieved by T13. In the end, Track 1 has a sensitivity of 22 mJy beam−1 per 0.1 $\mathrm{km}\ {{\rm{s}}}^{-1}$. In Track 2, the sensitivity was 30 mJy beam−1 per 0.1 $\mathrm{km}\ {{\rm{s}}}^{-1}$. A 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$integration results in a noise level of 21 mJy beam−1 $\mathrm{km}\ {{\rm{s}}}^{-1}$. The continuum sensitivity we achieve now is 0.22 mJy beam−1, compared with T13's value of 0.27 mJy per 2farcs× 2farcs0 beam.

3. CORE DETECTION ANALYSIS METHODS

Our main goal is to systematically identify ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores, eventually presenting a rank ordering of cores via their ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ line flux, which extends down to relatively weak cores—for which there begins to be possible confusion with noise fluctuations. We will utilize the information of collocation of the cores with other tracers (e.g., 1.3 mm continuum emission and other line tracers) to help assess the reliability of the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores. For the strongest cores, we will then estimate masses from both 1.3 mm dust emission and from the MIREX maps. Those cores with $\gtrsim 10\ {M}_{\odot }$ and without star-forming activity are good candidates to be massive starless cores.

We start by identifying cores as connected groups of "voxels" (pixels in 3D position–position–velocity (PPV) space of $l-b-v$) with flux densities of $\geqslant 3\sigma $ ("first threshold"), where σ is the noise level in each voxel. Note that this search is done in the cube before primary beam correction, to avoid spurious features at the edge of the maps; also, we restrict the core finding to within a field of view defined by the primary beam diameter. However, the noise level, σ, depends on the velocity resolution to which we smooth the data. After some experimentation (discussed below), we settle on a choice of a velocity resolution of 0.15 $\mathrm{km}\ {{\rm{s}}}^{-1}$ (i.e., three times coarser than the full resolution of the data cubes), which has $\sigma \simeq 0.020\ \mathrm{Jy}\ {\mathrm{bm}}^{-1}\ $ for IRDCs A, B, C, and E, (Track 1) and $\sigma \simeq 0.030\ \mathrm{Jy}\ {\mathrm{bm}}^{-1}\ $ for IRDCs D, F, and H (Track 2). Next, searching only within the FOV of diameter of 26'' (i.e., the half-power response diameter of the primary beam, i.e., where the sensitivity is reduced by about 50%), we identify all voxels above the $3\sigma $ threshold.

The selected voxels are then denoted as "nodes" (as in Graph theory). If two nodes are connected, i.e., adjacent in $l-b-v$ space, they are connected with an "edge." Once the data cube is traversed, we have a set of nodes and edges—and thus, a graph. Note that the graph so constructed has no directional information (i.e., an "undirected graph"): if voxel A is connected to voxel B, so is voxel B to voxel A. Next, using the NetworkX package (Hagberg et al. 2008), we identify all connected components in the graph, each of them being a core candidate.

After checking the results and comparing with the region (clump-scale) integrated intensity maps of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) (using a 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$velocity range), we notice some potential ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) structures that are not found by the above selection method. These structures have very few voxels that are above the first core threshold. However, they show continuous positive flux in velocity space. This leads us to amend our voxel selection: for already identified $\geqslant 3\sigma $ voxels, we consider their neighboring voxels and count them as part of the core, i.e., as nodes in the Graph, if they have a positive ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line flux. The effect of this is to add the contribution of a halo of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line emission around the stronger peaks.

One can see that core selection may depend on a number of choices of criteria, including PPV cube velocity resolution, noise selection thresholds, and minimum angular size. We considered a variety of velocity resolutions with which to smooth the data cubes, which have an original full resolution of 0.05 $\mathrm{km}\ {{\rm{s}}}^{-1}$. Based on the observations of T13, ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores show line widths (including hyperfine broadening) as narrow as $\sim 0.5\ \mathrm{km}\ {{\rm{s}}}^{-1}$. Therefore, although we have adopted a smoothing of the data to 0.15 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$resolution, we also make it a requirement that the core is detected in at least two adjacent channels. Because the ALMA data, by default, is Hanning smoothed, two neighboring channels are potentially correlated. Thus, a noise spike could potentially span across two channels, even when smoothed to 0.15 $\mathrm{km}\ {{\rm{s}}}^{-1}$, causing a false detection. To eliminate such false cores, we re-make the PPV cube with a 0.075 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$velocity shift in the boundaries of the velocity channels, and require cores to also be identified in this cube. The 0.075 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$shift is more than twice the finest channel spacing of the raw data, so these two cubes provide a good cross check to rule out noise spike features. Finally, we impose a condition of a minimum number of voxels that span an area on the sky that is comparable to the synthesized beam. This condition also helps mitigate false detections due to noise spikes.

Once ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores have been identified, we make additional checks of other tracers co-located at the core position and velocity. At the position of an ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ core, we check the signal in the continuum image and the 0th-moment images of DCO+(3-2) and C18O(2-1). Based on the observational results from T13, all ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ cores have corresponding ${\mathrm{DCO}}^{+}\,$and continuum emission. In cold starless cores, CO and its isotopologues should suffer heavy depletion via freeze-out on to dust grains, thus significantly reducing the flux in, e.g., C18O(2-1). However, note that these checks are not used in core selection, but only to provide additional information that can then be used to assess the reality of the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ core.

Finally, having obtained our sample of selected ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores, we re-examine their correspondence in the region (clump-scale) integrated intensity maps of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) that use a 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$velocity range (thus, with a higher noise level). The strongest cores are easily visible in this map, but many of the weaker ones do not stand out above the noise. In addition, there remain some $\geqslant 4\sigma $ features visible in the region maps that have not been identified as cores by our above method (i.e., as connected structures in PPV space). Because our focus is on the cores, we simply note the position of these "clump-scale" structures and carry out simple checks with other tracers to gauge their reality.

Our adopted method of finding cores, like any method, does involve some somewhat arbitrary choices for core definition and extraction. Thus, we have compared it to the "clumpfind" method in the yt package (Smith et al. 2009). We carried out the comparison in the B1 region. If we set the minimum size as the synthesized beam size, the search radius as the primary beam, and the threshold as 2.5σ, then yt finds all B1 cores given by our method. We consider that both methods find the coherent signals in the PPV cube. Ultimately, though, the precise definition used for a core will influence the final results. Thus, when comparing, for example, to the outputs of numerical simulations, similar analysis methods should be adopted where possible.

4. RESULTS

4.1. Overview of the 32 IRDC Clumps

Figures 1 through 10 present summary maps of the 30 targets, with one column for each source. Note that, except in the cases of D2 and D5, the FOV is centered on the IRDC core/clump Σ peak from BT12. The D2 region includes both D2 and (part of) D4, whereas the D5 region includes both D5 and D7. In these cases, the FOV is centered in between the Σ peaks. In all panels, the colored background shows MIREX-estimated mass surface density, Σ, in ${\rm{g}}\ {\mathrm{cm}}^{-2}$, from BT12. All panels also show the primary beam diameter FOV, which delimits the region that is searched for ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores. The locations ("+" signs) and names (XXA, XXB, XXC..., indicating the ordering of decreasing ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line flux) of these cores are also shown. These cores are discussed below, in Section 4.2. The panels also indicate the locations ("×" signs) of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) "clump-scale" structures with $\geqslant 4\sigma $ peaks in the 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$velocity range 0th-moment map that are not found by our ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) core detection algorithm.

Figure 1.

Figure 1. Summary figures for the surveyed IRDC clumps. Columns from left to right show the results for sources A1, A2, A3 (other sources continue in subsequent figures). The color-scale background shows MIREX mass surface density map from BT12. The color bar at the top-right corner indicates the scale in g cm−2, which is preserved throughout all 30 columns. The white regions are locations of MIR bright sources, where the map is undefined. The ALMA survey results are shown as contours. From top to bottom, the contours show 1.30 mm continuum, integrated intensities of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2), DCO+(3-2), DCN(3-2), ${{\rm{C}}}^{18}{\rm{O}}$(2-1), ${\mathrm{CH}}_{3}\mathrm{OH}$(v t = 0 5(1,4)-4(2,2)). The contour levels are in unit of σ, starting from 3σ, 4σ, 5σ, 7σ, 10σ, 15σ, 20σ, 30σ, 40σ, 50σ, 70σ, 100σ, 130σ, 160σ, 190σ... The σ for 1.30 mm continuum is 2.2 × 10−4 $\mathrm{Jy}\ {\mathrm{bm}}^{-1}$. The integrated intensity maps are made within a 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$velocity range, with a σ of about 0.02 $\mathrm{Jy}\ {\mathrm{bm}}^{-1}\ $ $\mathrm{km}\ {{\rm{s}}}^{-1}$. In each panel, the diameter of the primary beam is shown with the dashed circle and the ALMA beam on the lower left, and Spitzerμm beam on the lower right (which sets the resolution of the MIREX map). Each panel also shows locations of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores identified as connected structures in PPV space with "+" signs, along with the core names, e.g., A1A. Note that sometimes these cores do not appear as $\geqslant 3\sigma $ features in the integrated intensity map of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) in the second row (see text). Sometimes $\geqslant 4\sigma $ features that are seen in this map are not associated with identified cores, and these are marked with "×" signs (see text).

Standard image High-resolution image
Figure 2.

Figure 2. Same as Figure 1, but for sources B1, B2, and C2.

Standard image High-resolution image
Figure 3.

Figure 3. Same as Figure 1, but for sources C3, C4, and C5.

Standard image High-resolution image
Figure 4.

Figure 4. Same as Figure 1, but for sources C6, C7, and C8.

Standard image High-resolution image
Figure 5.

Figure 5. Same as Figure 1, but for sources C9, E1, and E2.

Standard image High-resolution image
Figure 6.

Figure 6. Same as Figure 1, but for sources D1, D2, and D3. Note that the center of the D2 core/clump from BT12 is shown with an open circle. The D4 complex is at the top of the D2 box (i.e., only partially within the 26'' diameter primary beam FOV), and the center of the D4 core/clump from BT12 is just outside the displayed box. For naming purposes, we assign all identified cores to D2.

Standard image High-resolution image
Figure 7.

Figure 7. Same as Figure 1, but for sources D5, D6, and D8. Note that the D5 box also contains D7, and the centers of these core/clumps from BT12 are shown with open circles.

Standard image High-resolution image
Figure 8.

Figure 8. Same as Figure 1, but for sources D9, F3, and F4.

Standard image High-resolution image
Figure 9.

Figure 9. Same as Figure 1, but for sources H1, H2, and H3.

Standard image High-resolution image
Figure 10.

Figure 10. Same as Figure 1, but for sources H4, H5, and H6.

Standard image High-resolution image

In the top row, the contours show 1.3 mm continuum emission. All regions generally have very strong continuum detection (C3 only has a 3σ detection, while D2 has 4σ detection). There is often quite good correspondence between the 1.3 mm continuum structures and those seen in the BT12 MIREX maps. Note, however, that the mm continuum image is insensitive to the structures that are larger than the maximum recoverable scale of the observation (i.e., ≳20''), unlike the MIREX map. Still, there are many regions where a MIREX Σ peak is not an especially strong mm continuum source and vice versa. Such discrepancies may arise because of problems in the MIREX map (e.g., where MIR bright sources are present) and/or if the mm continuum emission is being enhanced by higher temperatures, e.g., from local protostellar heating. A lack of prominent mm continuum emission from high Σ MIREX peaks may indicate these regions are extremely cold.

In the second row, the contours show ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) integrated intensity (0th-moment map). Here, and for all 0th-moment maps, the integration spans a velocity range of 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$centered on ${v}_{\mathrm{LSR}}$ of each IRDC9 (see Table 1). We find the strongest detections of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) in the B1, C9, and H2 clumps, which all show extended structures that are also identified as some of the highest ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line flux cores. In general, the strongest cores in a given region (e.g., A1A, A3A, etc) are easily visible in these 0th-moment maps, but many of the weaker cores do not stand out above the noise (lowest contour is 3σ ≃ 60 mJy bm−1 $\mathrm{km}\ {{\rm{s}}}^{-1}$), which is higher than the noise resulting from the more localized velocity ranges used to detect the cores.

In addition, there remain some $\geqslant 4\sigma $ features visible in these region maps that have not been identified as cores (i.e., as connected structures in PPV space). Note, however, that our core search is confined to within the primary beam, so strong features beyond this scale would not have been identified. Because our focus is on the cores, we only note here the position of these "clump-scale" structures (on average, there are about two per region), and discuss below their likelihood of being real features or merely noise fluctuations, by comparison with other tracers.

The third to sixth rows show 0th-moments maps of DCO+(3-2), DCN(3-2), ${{\rm{C}}}^{18}{\rm{O}}$(2-1), and CH3OH(v t = 0 5(1,4)-4(2,2)), respectively. These images contain a lot of ancillary information, which will not be discussed in detail in this paper. We note that all regions have strong ${{\rm{C}}}^{18}{\rm{O}}$(2-1) detection, which tends to trace spatially extended structures (again, subject to the constraints of spatial filtering of the largest scales). The other tracers tend to identify smaller scale structures.

One particular use of these images is to show the presence or absence of these species at the locations of identified ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores and "clump-scale" structures. Cores will be discussed in more detail in Section 4.2. For the clump-scale structures, several are also seen in these ancillary tracers, e.g., in: DCO+(3-2), such as the source in A3; DCN(3-2), such as the MIR-bright source in B2; or 1.3 mm continuum, such as sources in C6 (north) and F3 (also seen in CH3OH). When such coincident detections in independent tracers are seen, it increases our confidence in the reality of the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) detected structure. For the remainder, it is hard to be sure about their reality: more sensitive follow-up observations are needed. There may be several reasons these ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) structures are not identified as cores: they may not be well-connected in velocity space; they may not subtend a sufficient angular area compared to the beam; they may be too close to the boundary of the primary beam, where we truncate the core search algorithm; or they may already be attached to an identified core, but have their clump-scale 0th-moment peak slightly displaced from the core peak (e.g., near H2J and H3A).

Physically, some of the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) structures that are only seen in the broader, 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$velocity range 0th-moment map may represent an early stage of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) core formation, i.e., when the material is less concentrated in position–velocity space. Thus, the follow-up of such structures with higher sensitivity observations is warranted to investigate such a possibility.

4.2. Identified N2D+ Cores

In total, 141 ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores are found in the 30 data cubes, based on the criteria stated in Section 3. They are labeled as + signs in the second row of Figures 1 through 10. The cores are named based on the ranking of their ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) flux within each surveyed region: e.g., A3A is the highest-flux core within the A3 region, followed by A3B, A3C, etc.

Most strong features in the 0th-moment maps are identified, along with many cores that do not show up in these maps. Note, these 0th-moment maps are integrated over a 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$range. Meanwhile, many weak ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ cores are typically found in a velocity range much less than 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$(the narrowest velocity range is 0.30 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$by definition, corresponding to two channels, see Section 3). As a result, many of the weak cores do not show up in the region figures, i.e., at the positions of these cores there is no corresponding ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ contours seen.

We have ranked the cores based on their ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ 0th-moment flux, and list the strongest 50 of these in Table 2, along with the six cores from T13. The table lists core positions (intensity-weighted center) in Galactic coordinates in columns 2 and 3. Columns 4 and 5 give the velocity range over which the core is detected. Note the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cube used for core finding has been smoothed to 0.15 $\mathrm{km}\ {{\rm{s}}}^{-1}$ velocity resolution. In general, cores with structures in more channels tend to be more significant. Column 6 shows the intensity-weighted ${v}_{\mathrm{LSR}}$. and Column 7 shows the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) total line flux, ${S}_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}$, for each core. Columns 8, 9, and 10 show fluxes at the core position in the continuum image, the ${\mathrm{DCO}}^{+}\,$(3-2) 0th-moment image, and the ${{\rm{C}}}^{18}{\rm{O}}\,$(2-1) 0th-moment image (note that the T13 spectral set-up did not have ${{\rm{C}}}^{18}{\rm{O}}\,$(2-1)). The 0th-moment images of ${\mathrm{DCO}}^{+}\,$and ${{\rm{C}}}^{18}{\rm{O}}\,$are integrated within the velocity ranges shown in columns 4 and 5. The relevant signal-to-noise ratio (S/N) is also listed in these columns. Note that sometimes these fluxes can be negative in the case of non-detections where noise or other fluctuations cause the signal to be negative.

Table 2.  ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) Cores

Core l b ${v}_{\min }$ ${v}_{\max }$ $\bar{v}$ ${S}_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}$ ${S}_{1.30\mathrm{mm}}$/S/N ${S}_{{\mathrm{DCO}}^{+}}$/S/N ${S}_{{{\rm{C}}}^{18}{\rm{O}}}$/S/N
  (deg) (deg) ($\mathrm{km}\ {{\rm{s}}}^{-1}$) ($\mathrm{km}\ {{\rm{s}}}^{-1}$) ($\mathrm{km}\ {{\rm{s}}}^{-1}$) (Jy $\mathrm{km}\ {{\rm{s}}}^{-1}$) ($\mathrm{Jy}\ {\mathrm{bm}}^{-1}\,$ $\mathrm{km}\ {{\rm{s}}}^{-1}$) ($\mathrm{Jy}\ {\mathrm{bm}}^{-1}\,$ $\mathrm{km}\ {{\rm{s}}}^{-1}$) ($\mathrm{Jy}\ {\mathrm{bm}}^{-1}\,$ $\mathrm{km}\ {{\rm{s}}}^{-1}$)
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)
1 C9A 28.39894 0.08112 77.05 80.35 78.44 1.4e+00 6.9e−03/30 −8.6e−03/−0.88 −2.3e−01/−21
C1S 28.32194 0.06737 76.90 81.90 79.40 6.3e−01 1.6e−02/59 6.7e−02/8.2 ⋯/⋯
G2N 34.78101 −0.56816 38.95 43.95 41.45 6.1e−01 1.4e−03/8.3 1.5e−01/14 ⋯/⋯
2 B1A 19.28744 0.08048 26.40 29.10 27.66 4.6e−01 3.4e−04/1.5 4.4e−02/5.0 2.6e−02/2.7
C1N 28.32508 0.06714 78.68 83.68 81.18 2.7e−01 1.9e−03/6.9 4.6e−02/5.6 ⋯/⋯
3 H2A 35.48231 −0.28684 44.95 47.20 45.60 2.3e−01 5.4e−04/2.4 4.7e−02/4.7 2.7e−02/2.3
G2S 34.77842 −0.56838 39.30 44.30 41.80 2.2e−01 1.2e−03/6.7 1.1e−01/10 ⋯/⋯
F1 34.41928 0.24588 53.62 58.62 56.12 1.6e−01 2.9e−03/8.7 1.0e−01/12 ⋯/⋯
F2 34.43525 0.24140 55.16 60.16 57.66 9.5e−02 3.5e−03/16 1.5e−01/16 ⋯/⋯
4 B1B 19.28583 0.08304 26.40 27.15 26.76 6.5e−02 7.6e−03/23 3.3e−02/7.0 6.7e−03/1.3
5 C9B 28.39679 0.08314 76.45 77.95 77.30 6.5e−02 1.1e−04/0.32 −1.2e−02/−1.7 −5.4e−02/−7.5
6 H2B 35.48377 −0.28801 45.10 46.90 46.10 6.3e−02 2.6e−03/11 2.5e−02/2.7 4.1e−02/3.9
7 D6A 28.55294 −0.24057 85.50 86.85 86.39 6.0e−02 3.5e−04/0.93 −1.0e−02/−1.2 1.1e−02/1.2
8 D5A 28.56350 −0.23050 87.90 88.95 88.28 5.8e−02 4.3e−04/1.7 2.6e−03/0.33 −2.3e−02/−2.9
9 B2A 19.30983 0.06697 24.90 26.55 25.56 5.6e−02 2.8e−03/9.8 2.8e−02/4.2 5.5e−02/7.5
10 H5A 35.49591 −0.28680 45.10 46.30 45.34 5.2e−02 4.1e−04/1.7 3.3e−02/3.9 2.0e−02/2.2
11 H1A 35.47838 −0.30785 44.20 45.55 44.88 5.0e−02 −2.2e−04/−0.68 2.5e−02/3.0 2.3e−02/2.5
12 H2C 35.48314 −0.28847 44.95 46.45 45.69 4.9e−02 9.0e−04/3.6 2.1e−02/2.6 3.0e−02/3.1
13 F4A 34.45842 0.25645 56.95 58.15 57.40 4.9e−02 1.4e−03/6.4 4.7e−02/6.0 1.5e−02/1.7
14 C3A 28.34951 0.09583 77.35 79.00 78.29 4.9e−02 3.8e−04/1.1 6.7e−03/1.0 −1.7e−02/−2.4
15 C5A 28.35517 0.05754 76.45 77.65 76.80 4.8e−02 −7.0e−05/−0.29 4.0e−03/0.72 1.4e−03/0.22
16 D2A 28.53882 −0.27327 84.45 85.80 84.73 4.8e−02 1.2e−04/0.40 4.0e−04/0.048 −9.0e−03/−0.94
17 H2D 35.48285 −0.28575 44.35 47.20 45.68 4.6e−02 2.8e−04/1.2 2.7e−02/2.4 −5.6e−02/−4.3
18 H2E 35.48511 −0.28485 45.55 46.60 46.08 4.6e−02 1.9e−04/0.51 5.4e−02/7.5 3.5e−04/0.044
19 D2B 28.53645 −0.27670 85.50 86.55 86.11 4.6e−02 1.3e−05/0.047 −1.0e−02/−1.3 −2.5e−03/−0.31
20 C4A 28.35450 0.07184 79.15 80.65 79.87 4.4e−02 5.0e−04/2.2 8.5e−03/1.4 2.3e−02/3.3
21 C2A 28.34592 0.05881 77.35 79.75 78.36 4.3e−02 5.9e−04/1.9 8.1e−03/0.94 −1.6e−02/−1.7
22 C2B 28.34677 0.05918 79.45 79.60 79.50 4.1e−02 2.4e−03/6.5 −3.1e−03/−0.92 −4.9e−03/−1.8
23 A3A 18.80730 −0.30530 64.10 66.05 65.16 3.9e−02 1.8e−03/8.2 2.4e−02/3.3 1.8e−02/2.4
24 D1A 28.52688 −0.25208 87.30 87.90 87.46 3.8e−02 1.3e−04/0.50 1.1e−02/1.8 3.0e−02/4.5
25 A3B 18.80692 −0.30445 63.95 65.90 65.15 3.7e−02 4.3e−05/0.18 1.5e−02/2.1 2.3e−02/3.1
26 H1B 35.47858 −0.31096 43.90 45.40 44.49 3.6e−02 5.5e−04/2.4 9.3e−03/1.1 −8.1e−03/−0.84
27 D2C 28.53883 −0.27762 88.65 88.80 88.75 3.5e−02 1.9e−04/0.57 −3.3e−03/−0.79 4.3e−03/1.0
28 D1B 28.52548 −0.25163 86.40 86.70 86.58 3.5e−02 6.6e−04/2.4 3.3e−03/0.74 1.1e−05/2.1e−3
29 C7A 28.36254 0.12185 79.45 80.50 80.20 3.4e−02 2.8e−04/0.76 −2.5e−03/−0.42 −2.7e−02/−4.6
30 D7A 28.56324 −0.23270 84.45 84.90 84.65 3.3e−02 1.2e−04/0.40 −2.4e−03/−0.46 −1.1e−02/−1.8
31 D2D 28.54054 −0.27463 88.80 89.25 89.09 3.3e−02 −1.0e−04/−0.29 6.1e−04/0.12 −2.6e−03/−0.48
32 D8A 28.57531 −0.23401 84.45 85.50 84.65 3.3e−02 −2.5e−04/−0.78 1.3e−02/1.8 −2.7e−03/−0.32
33 D3A 28.54180 −0.23689 85.80 86.85 86.66 3.3e−02 2.8e−04/1.1 −3.3e−02/−4.5 1.6e−02/2.0
34 C4B 28.35479 0.07069 80.05 80.20 80.14 3.2e−02 9.8e−04/4.4 2.7e−02/8.6 8.2e−03/3.3
35 D9A 28.58601 −0.22902 85.95 87.45 86.63 3.2e−02 −8.4e−05/−0.34 2.1e−03/0.25 −4.8e−04/−0.049
36 D2E 28.53890 −0.27550 87.15 88.65 88.01 3.0e−02 −1.7e−04/−0.71 −5.9e−03/−0.71 −3.1e−02/−3.1
37 B2B 19.30633 0.06616 25.95 27.45 26.74 2.9e−02 1.2e−03/5.2 3.0e−02/4.4 3.1e−02/4.4
38 C9C 28.40155 0.07969 78.70 79.30 78.97 2.8e−02 −8.8e−04/−2.2 4.2e−03/0.92 −5.2e−03/−1.1
39 C4C 28.35621 0.07074 79.15 80.95 79.95 2.8e−02 −2.0e−05/−0.074 5.1e−03/0.72 1.2e−02/1.6
40 H6A 35.52262 −0.27226 44.95 46.15 45.49 2.7e−02 4.8e−03/22 5.0e−02/6.1 1.6e−01/18
41 D1C 28.52562 −0.25044 86.55 86.70 86.64 2.6e−02 3.0e−05/0.12 1.1e−03/0.36 −5.4e−03/−1.2
42 C2C 28.34311 0.06017 78.40 80.20 79.14 2.6e−02 8.8e−03/39 2.9e−02/3.8 1.3e−01/15
43 H2F 35.48063 −0.28927 45.10 46.60 46.02 2.5e−02 5.2e−04/1.4 −5.6e−04/−0.068 −2.4e−02/−2.5
44 H3A 35.48931 −0.29409 42.40 44.65 43.33 2.5e−02 2.9e−04/1.1 4.0e−02/3.9 −3.4e−03/−0.28
45 D7B 28.56577 −0.23133 89.10 89.25 89.24 2.5e−02 −1.4e−04/−0.62 2.7e−03/0.65 −7.8e−03/−2.3
46 C6A 28.36492 0.05084 80.20 81.25 80.63 2.4e−02 −3.2e−04/−1.1 1.2e−02/2.3 8.0e−03/1.4
47 H2G 35.48368 −0.28475 45.10 45.25 45.25 2.4e−02 3.2e−05/0.11 1.7e−02/5.9 6.4e−03/1.5
48 D3B 28.54043 −0.23394 87.60 88.65 87.83 2.4e−02 4.6e−04/1.9 1.5e−02/2.0 3.8e−02/4.7
49 D8B 28.57318 −0.23559 88.50 89.25 88.72 2.3e−02 1.2e−04/0.35 −2.4e−03/−0.36 −8.2e−03/−1.2
50 D2F 28.53729 −0.27641 87.30 87.45 87.40 2.3e−02 7.2e−04/3.0 −2.8e−03/−0.67 −4.0e−03/−1.2

Download table as:  ASCIITypeset image

Figures 1113 show the top 15 newly detected ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores, ordered by ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line flux. Each figure row shows a 10'' by 10'' zoom-in view of the core, centering at the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) intensity weighted center. The first panel in each row shows the 1.3 mm continuum contours, overlaid on the MIREX map (BT12). For all figures, we use the same color scale for the MIREX image.

Figure 11.

Figure 11. Core-scale zoom-in figures of C9A, B1A, H2A, B1B, C9B (rows top to bottom, also following rank order of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) flux). Each panel shows a 10'' by 10'' FOV. The first column shows 1.3 mm continuum (contours: 2, 3, 4, 5, 7, 9, 11, 13, 15, 17, 20, 40, 60, 80... σ, where $1\sigma \simeq 2.2\times {10}^{-4}\,\mathrm{Jy}\ {\mathrm{bm}}^{-1}$; ALMA beam in lower left) overlaid on MIREX (BT12) mass surface density (color scale in g cm−2; Spitzer beam in lower right). 2nd column shows 0th-moment ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) of the cores (only including core voxels), shown as the solid contours: 2, 3, 4, 5...σ (beam lower left), overlaid on the 1.30 mm continuum (beam lower right). The dotted contours are the full integration of the cores (including all voxels within the channels). At the top of the panel we list total ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) flux (in $\mathrm{Jy}\ {\mathrm{bm}}^{-1}\ $ $\mathrm{km}\ {{\rm{s}}}^{-1}$), ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) 0th-moment map rms (in $\mathrm{Jy}\ {\mathrm{bm}}^{-1}\ $ $\mathrm{km}\ {{\rm{s}}}^{-1}$), and ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) intensity weighted core velocity and velocity range (in $\mathrm{km}\ {{\rm{s}}}^{-1}$). The solid and dashed ellipses are fitted boundaries to the cores (see text). 3rd column is the same as the 2nd, except contours now show DCO+(3-2) integrated over the same velocity range as the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ core. 4th column is the same as the 2nd, except contours now show ${{\rm{C}}}^{18}{\rm{O}}$(2-1) integrated over the same velocity range as the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ core. 5th column is the same as the 2nd, except contours now show SiO(5-4) integrated over the 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$velocity range used in the region maps.

Standard image High-resolution image
Figure 12.

Figure 12. Same as Figure 11, but for H2B, D6A, D5A, B2A, H5A.

Standard image High-resolution image

The second panel shows the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$(3-2) integrated intensity (black solid contours) on top of the 1.3 mm continuum emission (same color scale for all sources). The integration for these ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$(3-2) 0th-moment maps is only over velocity channels in which the core is defined and only including voxels flagged to be in the core. We also show a full 0th-moment integration map as gray dotted contours, which includes all voxels within the velocity ranges. Both types of contour maps start at 2σ and increase with a step of 1σ, with σ being the noise of the full integration map. The gray dotted contours always extend over a wider area than the black ones because the full integration picks up more flux at the core boundary, which is also why the black solid contours are sometimes separated. At the top of the panel, several properties of the core: ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) flux, ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) 0th-moment map rms, and ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) intensity weighted core velocity (sub- and super-scripts are velocity boundaries over which core is detected).

The third, fourth and fifth panels show DCO+(3-2), ${{\rm{C}}}^{18}{\rm{O}}\,$(2-1) and SiO(5-4) 0th-moment maps on top of the dust continuum. For DCO+(3-2) and ${{\rm{C}}}^{18}{\rm{O}}\,$(2-1), the integrations are over the same velocity ranges as the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) 0th-moment map. For SiO(5-4) the velocity range is the 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$range used in the region maps (note a full analysis of the SiO(5-4) data from the region maps will be presented in a companion paper by M. Liu et al. 2016, in preparation).

Comparing some of the top ranked cores, including those of T13 (see Table 2), C9A has more than twice the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) flux of C1-S. B1A ranks above C1-N, and H2A ranks above F1. Note that the integration for C9A, B1B, H2A only includes voxels in the connected components (the defined cores), so their fluxes are higher in a full integration (i.e., gray dotted contours in these core figures). However, we also note that the sensitivity in this data set is about 3 times worse than T13, so the uncertainty in core fluxes is also higher.

One of the features of the 6 ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores found by T13 was their extended DCO+(3-2) envelopes. This is still true for most of the new cores presented here, however, now we also see some cores, like C9A, C9B, D5A, C3A and C5A, which have relatively weak DCO+(3-2) emission. This chemical diversity may reflect different environmental conditions among the core sample.

We note that there is often a dearth of ${{\rm{C}}}^{18}{\rm{O}}$(2-1) emission associated with the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores, which likely indicates that there is a high degree of gas phase depletion of CO, i.e., due to freeze out onto dust grain ice mantles. Such CO depletion is thought to boost the deuteration of ${{\rm{N}}}_{2}{{\rm{H}}}^{+}\,$(e.g., K15), and so this anti-correlation of CO emission with ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ emission is expected theoretically.

SiO(5-4) emission is a known tracer of protostellar outflows and so can help us assess such activity within the cores. C9A and C9B show extended SiO emission in their vicinity, which is however thought to mostly arise from a separate massive protostellar source (e.g., that is seen as a strong DCN(3-2) source in the lower right of the region Figure 5) (M. Liu et al. 2016, in preparation). However, there is some localized, potentially elongated SiO emission that is spatially coincident with C9A, which likely indicates that this core is already forming a protostar (similar to that present in C1-S Feng et al. 2016; Tan et al. 2016). Most other cores do not show strong SiO(5-4) emission. We discuss more details about the six strongest ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores below.

In order to study basic core properties and, for the best cases of C9A, B1A, H2A, B1B, B2A and F4A, compare to simple dynamical models, we fit 2D elliptical Gaussians to the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) integrated intensity maps. We define an ellipse at the $3\sigma $ value of the 2D Gaussian (Case 1), which is shown as the white, solid ellipses in the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) columns of Figures 11 to 13. These fitting results are listed in Table 3 for the best six cores. Note that the 2D Gaussian center is sometimes slightly different from the intensity weighted core center defined earlier (Table 2). Note also that the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) 0th-moment image, e.g., for C9A, has an rms of 13 mJy per 1farcs× 1farcs0 beam $\mathrm{km}\ {{\rm{s}}}^{-1}$, roughly corresponding to 23 mJy per 2farcs× 2farcs$\mathrm{km}\ {{\rm{s}}}^{-1}$, which is about twice the rms of C1 region rms reported by T13. Thus the core sizes reported here are likely to be systematically somewhat smaller than if they had been measured with the same sensitivity used by T13. Still, we will use these ellipses for our following study of core dynamics. For C9A, B1A, H2A, B1B, B2A and F4A we also consider their properties on the scale of larger ellipses (Case 2), shown by the dashed ellipses in Figures 1113). These geometries are based on total ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) intensities: they cover most of the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ flux within the 2σ contours from the full integration images. These ellipses also cover most of the 1.3 mm continuum flux that is seen to be associated with these cores. These core properties are shown inside square brackets in Table 3. We note that given these uncertainties in defining core sizes, we have not attempted to derive deconvolved sizes. As discussed above, our Case 1 radii are probably lower limits, and deconvolution would have only a very minor effect on the derived Case 2 radii.

Figure 13.

Figure 13. Same as Figure 11, but for H1A, H2C, F4A, C3A, C5A.

Standard image High-resolution image

Table 3.  Physical Properties of the Six "Best" ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ Cores

Core property (% error) C9A B1A H2A B1B B2A F4A Average
l (°) 28.39896 19.28747 35.48231 19.28584 19.30986 34.45843 ...
  [28.39896] [19.28747] [35.48231] [19.28567] [19.30983] [34.45846] ...
b (°) 0.08113 0.08050 −0.28681 0.08305 0.06689 0.25643 ...
  [0.08113] [0.08050] [−0.28681] [0.08308] [0.06699] [0.25651] ...
${\theta }_{c}$ ('') 1.94 1.65 1.42 0.707 0.745 0.606 ...
  [3.30] [3.05] [2.28] [1.44] [1.80] [1.53] ...
e 0.747 0.877 0.852 0.881 0.672 0.955 ...
  [0.904] [0.866] [0.781] [0.00] [0.00] [0.866] ...
P.A. (°) 16 42 45 58 83 85 ...
  [16] [20] [45] [0] [0] [70] ...
d (kpc) (20%) 5.0 2.4 2.9 2.4 2.4 3.7 ...
${R}_{{\rm{c}}}$ (0.01 pc) (20%) 4.71 1.92 2.00 0.822 0.867 1.09 ...
  [7.99] [3.55] [3.20] [1.67] [2.09] [2.74] ...
${V}_{\mathrm{LSR},{{\rm{N}}}_{2}{{\rm{D}}}^{+}}$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ 78.40 ± 0.02 27.60 ± 0.02 45.50 ± 0.02 26.60 ± 0.04 25.50 ± 0.03 57.10 ± 0.05 ...
  [78.40 ± 0.03] [27.50 ± 0.02] [45.50 ± 0.03] [26.70 ± 0.04] [25.50 ± 0.04] [57.10 ± 0.11] ...
${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+},\mathrm{obs}}$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ 0.579 ± 0.028 0.351 ± 0.025 0.245 ± 0.029 0.296 ± 0.068 0.180 ± 0.029 0.329 ± 0.053 ...
  [0.579 ± 0.028] [0.351 ± 0.025] [0.245 ± 0.029] [0.296 ± 0.068] [0.180 ± 0.029] [0.329 ± 0.053] ...
${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+},\mathrm{nt}}$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ 0.576 ± 0.028 0.347 ± 0.025 0.239 ± 0.030 0.291 ± 0.069 0.172 ± 0.030 0.325 ± 0.053 ...
  [0.576 ± 0.028] [0.347 ± 0.025] [0.239 ± 0.030] [0.291 ± 0.069] [0.172 ± 0.030] [0.325 ± 0.053] ...
${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ 0.606 ± 0.028 0.395 ± 0.026 0.304 ± 0.029 0.347 ± 0.062 0.255 ± 0.029 0.375 ± 0.049 ...
  [0.606 ± 0.028] [0.395 ± 0.026] [0.304 ± 0.029] [0.347 ± 0.062] [0.255 ± 0.029] [0.375 ± 0.049] ...
${{\rm{\Sigma }}}_{\mathrm{cl}}$ $({\rm{g}}\ {\mathrm{cm}}^{-2})$ (30%) 0.317 0.394 0.321 0.361 0.293 0.313 ...
  [0.312] [0.321] [0.292] [0.362] [0.273] [0.288] ...
${{\rm{\Sigma }}}_{{\rm{c}},\max }$ $({\rm{g}}\ {\mathrm{cm}}^{-2})$ (30%) 0.282 0.492 0.327 0.391 0.302 0.349 ...
  [0.308] [0.438] [0.326] [0.437] [0.297] [0.334] ...
${M}_{{\rm{c}},\max }$ $({M}_{\odot })$ (50%) 9.34 2.70 1.95 0.394 0.338 0.616 ...
  [29.3] [8.24] [4.99] [1.83] [1.94] [3.74] ...
${n}_{{\rm{H}},{\rm{c}},\max }({10}^{5}\,{\mathrm{cm}}^{-3})$ (36%) 6.17 26.5 16.9 49.1 35.9 33.1 ...
  [3.97] [12.7] [10.5] [26.9] [14.6] [12.6] ...
${\sigma }_{{\rm{c}},\mathrm{vir},\max }$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ 0.514 ± 0.064 0.398 ± 0.050 0.348 ± 0.044 0.241 ± 0.030 0.220 ± 0.027 0.259 ± 0.032 ...
  [0.681 ± 0.086] [0.500 ± 0.063] [0.430 ± 0.054] [0.353 ± 0.044] [0.334 ± 0.042] [0.399 ± 0.050] ...
${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}/{\sigma }_{{\rm{c}},\mathrm{vir},\max }$ 1.18 ± 0.14 0.992 ± 0.129 0.873 ± 0.130 1.44 ± 0.30 1.16 ± 0.18 1.45 ± 0.25 1.18 ± 0.08
  [0.890 ± 0.108] [0.790 ± 0.103] [0.707 ± 0.105] [0.982 ± 0.208] [0.763 ± 0.123] [0.941 ± 0.163] [0.845 ± 0.057]
${R}_{{\rm{c}},\mathrm{vir},\max }$ (0.01 pc) 4.02 ± 1.03 1.94 ± 0.49 1.83 ± 0.46 0.774 ± 0.198 0.796 ± 0.204 1.04 ± 0.26 ...
  [7.18 ± 1.84] [3.75 ± 0.96] [3.06 ± 0.78] [1.66 ± 0.42] [1.98 ± 0.50] [2.67 ± 0.68] ...
${R}_{c}/{R}_{{\rm{c}},\mathrm{vir},\max }$ 1.17 ± 0.33 0.989 ± 0.282 1.09 ± 0.31 1.06 ± 0.30 1.09 ± 0.31 1.05 ± 0.29 1.07 ± 0.12
  [1.11 ± 0.31] [0.946 ± 0.270] [1.05 ± 0.29] [1.01 ± 0.28] [1.06 ± 0.30] [1.03 ± 0.29] [1.03 ± 0.12]
${{\rm{\Sigma }}}_{{\rm{c}},\min }$ $({\rm{g}}\ {\mathrm{cm}}^{-2})$ (30%) - 0.0979 6.37e−3 0.0303 8.68e−3 0.0368 ...
  - [0.117] [0.0342] [0.0743] [0.0243] [0.0457] ...
${M}_{{\rm{c}},\min }$ $({M}_{\odot })$ (50%) - 0.537 0.0380 0.0305 9.72e−3 0.0650 ...
  - [2.20] [0.522] [0.311] [0.159] [0.511] ...
${n}_{{\rm{H}},{\rm{c}},\min }({10}^{5}\,{\mathrm{cm}}^{-3})$ (36%) - 5.26 0.329 3.80 1.03 3.49 ...
  - [3.39] [1.10] [4.58] [1.20] [1.72] ...
${\sigma }_{{\rm{c}},\mathrm{vir},\min }$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ - 0.266 ± 0.033 0.130 ± 0.016 0.127 ± 0.016 0.0905 ± 0.0114 0.148 ± 0.018 ...
  - [0.359 ± 0.045] [0.245 ± 0.030] [0.227 ± 0.028] [0.179 ± 0.022] [0.243 ± 0.030] ...
${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}/{\sigma }_{{\rm{c}},\mathrm{vir},\min }$ - 1.49 ± 0.19 2.34 ± 0.34 2.74 ± 0.57 2.82 ± 0.45 2.54 ± 0.43 2.38 ± 0.18
  - [1.10 ± 0.14] [1.24 ± 0.18] [1.53 ± 0.32] [1.43 ± 0.23] [1.55 ± 0.26] [1.37 ± 0.10]
${R}_{{\rm{c}},\mathrm{vir},\min }$ (0.01 pc) - 0.865 ± 0.222 0.255 ± 0.065 0.215 ± 0.055 0.135 ± 0.034 0.338 ± 0.086 ...
  - [1.94 ± 0.49] [0.991 ± 0.254] [0.687 ± 0.176] [0.565 ± 0.145] [0.987 ± 0.253] ...
${R}_{c}/{R}_{{\rm{c}},\mathrm{vir},\min }$ - 2.22 ± 0.63 7.84 ± 2.24 3.82 ± 1.09 6.42 ± 1.83 3.22 ± 0.92 4.70 ± 0.65
  - [1.83 ± 0.52] [3.23 ± 0.92] [2.44 ± 0.69] [3.71 ± 1.06] [2.78 ± 0.79] [2.80 ± 0.36]
${S}_{1.30\mathrm{mm}}$ (mJy) 31.3 ± 0.8 9.63 ± 0.69 5.52 ± 0.61 10.3 ± 0.3 4.36 ± 0.31 1.96 ± 0.26 ...
  [76.4 ± 1.3] [20.1 ± 1.2] [10.7 ± 0.9] [27.6 ± 0.6] [14.2 ± 0.7] [7.24 ± 0.66] ...
${S}_{1.30\mathrm{mm}}/{\rm{\Omega }}$ (MJy/sr) 112 ± 2 47.6 ± 3.4 37.1 ± 4.1 274 ± 7 107 ± 7 71.9 ± 9.6 ...
  [95.0 ± 1.7] [29.1 ± 1.8] [28.2 ± 2.5] [183 ± 3] [59.3 ± 3.1] [41.5 ± 3.8] ...
${{\rm{\Sigma }}}_{{\rm{c}},\mathrm{mm}}$ $({\rm{g}}\ {\mathrm{cm}}^{-2})$ ${2.10}_{1.24}^{4.21}$ ${0.895}_{0.530}^{1.79}$ ${0.698}_{0.413}^{1.40}$ ${5.15}_{3.05}^{10.3}$ ${2.01}_{1.19}^{4.03}$ ${1.35}_{0.800}^{2.71}$ ...
  [1.79${}_{1.06}^{3.58}]$ [0.547${}_{0.324}^{1.10}]$ [0.531${}_{0.314}^{1.06}]$ [3.44${}_{2.03}^{6.89}]$ [1.12${}_{0.660}^{2.23}]$ [0.781${}_{0.462}^{1.56}]$ ...
${M}_{{\rm{c}},\mathrm{mm}}$ $({M}_{\odot })$ ${69.7}_{31.7}^{146}$ ${4.91}_{2.24}^{10.3}$ ${4.16}_{1.90}^{8.71}$ ${5.19}_{2.36}^{10.9}$ ${2.26}_{1.03}^{4.73}$ ${2.39}_{1.09}^{5.00}$ ...
  [170${}_{77.6}^{357}]$ [10.3${}_{4.69}^{21.6}]$ [8.11${}_{3.70}^{17.0}]$ [14.4${}_{6.55}^{30.1}]$ [7.29${}_{3.32}^{15.3}]$ [8.73${}_{3.98}^{18.3}]$ ...
${n}_{{\rm{H}},{\rm{c}},\mathrm{mm}}$ $({10}^{5}\,{\mathrm{cm}}^{-3})$ ${45.9}_{25.5}^{93.8}$ ${48.0}_{26.6}^{98.1}$ ${35.9}_{19.9}^{73.4}$ ${644}_{357}^{1320}$ ${239}_{133}^{489}$ ${128}_{71.0}^{261}$ ...
  [23.0${}_{12.8}^{47.0}]$ [15.8${}_{8.79}^{32.4}]$ [17.1${}_{9.47}^{34.9}]$ [211${}_{117}^{432}]$ [54.8${}_{30.4}^{112}]$ [29.3${}_{16.3}^{59.9}]$ ...
${\sigma }_{{\rm{c}},\mathrm{vir},\mathrm{mm}}$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ ${0.849}_{0.681}^{1.03}$ ${0.462}_{0.371}^{0.561}$ ${0.421}_{0.338}^{0.511}$ ${0.458}_{0.368}^{0.556}$ ${0.353}_{0.283}^{0.429}$ ${0.364}_{0.292}^{0.442}$ ...
  [1.06${}_{0.849}^{1.28}]$ [0.528${}_{0.424}^{0.641}]$ [0.486${}_{0.390}^{0.590}]$ [0.592${}_{0.475}^{0.718}]$ [0.465${}_{0.373}^{0.565}]$ [0.493${}_{0.396}^{0.599}]$ ...
${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}/{\sigma }_{{\rm{c}},\mathrm{vir},\mathrm{mm}}$ ${0.714}_{0.588}^{0.890}$ ${0.854}_{0.703}^{1.06}$ ${0.723}_{0.595}^{0.901}$ ${0.758}_{0.624}^{0.944}$ ${0.722}_{0.594}^{0.900}$ ${1.03}_{0.850}^{1.29}$ ${0.800}_{0.742}^{0.882}$
  [0.573${}_{0.472}^{0.715}]$ [0.747${}_{0.615}^{0.931}]$ [0.626${}_{0.516}^{0.781}]$ [0.586${}_{0.483}^{0.731}]$ [0.548${}_{0.451}^{0.683}]$ [0.761${}_{0.627}^{0.948}]$ [0.640${}_{0.594}^{0.705}]$
${R}_{{\rm{c}},\mathrm{vir},\mathrm{mm}}$ (0.01 pc) ${11.0}_{6.82}^{16.1}$ ${2.62}_{1.62}^{3.83}$ ${2.67}_{1.66}^{3.91}$ ${2.81}_{1.74}^{4.11}$ ${2.06}_{1.28}^{3.01}$ ${2.05}_{1.27}^{3.00}$ ...
  [17.3${}_{10.7}^{25.3}]$ [4.20${}_{2.61}^{6.14}]$ [3.90${}_{2.42}^{5.72}]$ [4.67${}_{2.90}^{6.83}]$ [3.83${}_{2.38}^{5.61}]$ [4.08${}_{2.53}^{5.97}]$ ...
${R}_{c}/{R}_{{\rm{c}},\mathrm{vir},\mathrm{mm}}$ ${0.429}_{0.288}^{0.657}$ ${0.733}_{0.493}^{1.12}$ ${0.749}_{0.504}^{1.15}$ ${0.293}_{0.197}^{0.448}$ ${0.421}_{0.283}^{0.646}$ ${0.531}_{0.357}^{0.814}$ ${0.526}_{0.452}^{0.646}$
  [0.462${}_{0.310}^{0.708}]$ [0.846${}_{0.569}^{1.30}]$ [0.820${}_{0.551}^{1.26}]$ [0.359${}_{0.241}^{0.550}]$ [0.547${}_{0.367}^{0.838}]$ [0.672${}_{0.451}^{1.03}]$ [0.617${}_{0.531}^{0.757}]$

Download table as:  ASCIITypeset images: 1 2

In the following sections, we study the core dynamics, following the methods of T13. We focus on 6 cores C9A, B1A, H2A, B1B, B2A, and F4A. The first four of these are the top four in terms of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) flux of the new cores presented in this paper (see Table 2). They all have clear associations with 1.3 mm continuum structures, which will be the preferred method of estimated masses. B2A and F4A have relatively weak ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) fluxes, but also show quite good correspondence with continuum sources.

4.3. Core Kinematics

Only the top four ranked ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores are significantly larger than the beam. For these, Figure 14 shows the first-moment maps of this species, along with DCO+(3-2) and ${{\rm{C}}}^{18}{\rm{O}}$(3-2). Only voxels with $\geqslant 3\sigma $ detection are considered. C9A shows a strong velocity gradient in ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2): in the upper left part of the image, the mean velocities are $\simeq -1.5\mathrm{km}\ {{\rm{s}}}^{-1}$, whereas they are $\simeq +0.5\ \mathrm{km}\ {{\rm{s}}}^{-1}$ on the lower right. The core diameter is about 0.1 pc, so the velocity gradient is ∼20 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$ pc−1. Given the bimodal morphology of the 0th-moment map, it is possible that we are seeing two ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores in the process of merging. The other cores, B1A and H2A, do not show such large velocity gradients.

Figure 14.

Figure 14. First-moment (intensity-weighted ${v}_{\mathrm{LSR}}$) maps of the top four ranked ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores C9A (first row), B1A (second row), H2A (third row), B1B (fourth row), in ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$(first column), ${\mathrm{DCO}}^{+}\,$(second column), ${{\rm{C}}}^{18}{\rm{O}}\,$(third column). The calculation is within a 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$range centered on ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) ${v}_{\mathrm{LSR}}$ (Table 2). Only voxels with ≥3σ intensity are considered and shown in color. The color scale is linear from −2.5 to +2.5 $\mathrm{km}\ {{\rm{s}}}^{-1}$. The synthesized beam is shown in the lower-right corner.

Standard image High-resolution image

Figure 15 shows the second-moment (velocity dispersion) maps of C9A, B1A and H2A in ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2), DCO+(3-2) and ${{\rm{C}}}^{18}{\rm{O}}$(3-2). Again, only voxels with $\geqslant 3\sigma $ detection are considered. A 0.242 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$effective dispersion from the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) main hyperfine group is subtracted in quadrature from its maps. Regions that then have a negative result are shown as blank; here, the sensitivity is probably too low to obtain a good measure of the velocity dispersion. Note also this effect leads to the appearance of artificially low dispersion halos around the main features. The C9A ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) second-moment map shows a higher velocity dispersion on its left-hand side than its right. Again, this may be evidence in favor of the two merging cores scenario.

Figure 15.

Figure 15. Second-moment (intensity-weighted velocity dispersion) maps of the top four ranked ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores C9A (first row), B1A (second row), H2A (third row), B1B (fourth row), in ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$(first column), ${\mathrm{DCO}}^{+}\,$(second column), ${{\rm{C}}}^{18}{\rm{O}}\,$(third column). The calculation is within a 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$range, centered on ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,{v}_{\mathrm{LSR}}$ (Table 2). Only voxels with $\geqslant $ 3σ intensity are considered and shown in color. For the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ data, a 0.242 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$contribution to the dispersion from the main group of hyperfine structures is subtracted in quadrature. However, we see in some regions the second-moment pixels shrink from the first-moment images. This is mainly due to the fact that the velocity span is narrower than 0.242 $\mathrm{km}\ {{\rm{s}}}^{-1}$. In these areas, we are limited by the sensitivity. The color scale is linear from 0 to 1.0 $\mathrm{km}\ {{\rm{s}}}^{-1}$. The synthesized beam is shown in the lower-right corner.

Standard image High-resolution image

Within the defined Case 1 ellipses of C9A, B1A, H2A, B1B, B2A, and F4A, we extract total flux density from each channel (full velocity resolution of 0.04 $\mathrm{km}\ {{\rm{s}}}^{-1}$) and convert it to main beam temperature, ${T}_{\mathrm{mb}}$, to derive the total spectrum as a function of ${v}_{\mathrm{LSR}}$. These are shown in Figure 16. We then fit the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line with full blended hyperfine components, utilizing the HFS method of the CLASS software package,10 assuming the line is optically thin (this was shown to be good approximation in the cores studied by T13). The resulting centroid velocity ${V}_{\mathrm{LSR},{{\rm{N}}}_{2}{{\rm{D}}}^{+}}$ and 1D velocity dispersion ${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+},\mathrm{obs}}$ are listed in Table 3. The velocity dispersions derived by this method range from about 0.2 to 0.6 $\mathrm{km}\ {{\rm{s}}}^{-1}$. However, to estimate the total velocity dispersion that is needed for the dynamical analysis, we subtract the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ thermal component from its velocity dispersion in quadrature (assuming 10 K temperature), and add back the core sound speed in quadrature, assuming a mean particle mass $\mu =2.33{m}_{p}$ and gas temperature of 10 ± 3 K, following T13. These results, listed as ${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}$ in Table 3, are slightly larger, ranging from 0.26 to 0.61 $\mathrm{km}\ {{\rm{s}}}^{-1}$.

Figure 16.

Figure 16.  ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) core spectra (black) and their HFS fit results (blue). All spectra are in the rest frame of their centroid velocity ${V}_{\mathrm{LSR},{{\rm{N}}}_{2}{{\rm{D}}}^{+}}$, derived from the HFS fit. The core names are labeled on the top left. Note that the spectra shown are binned to have 0.08 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$spectral resolution, whereas the HFS fit results are drawn from their maximum velocity resolution of 0.04 $\mathrm{km}\ {{\rm{s}}}^{-1}$.

Standard image High-resolution image

4.4. Core Masses

Following the methods of T13, we measure core masses, Mc, within the Case 1 and Case 2 ellipses in two ways. First, we use the MIREX map of BT12. As a first MIREX-based estimate, we use the total mass surface density in the map integrated over the area of the core to yield a mass, ${M}_{c,\max }.$ As a second MIREX-based estimate, we account for the clump mass surface density, ${{\rm{\Sigma }}}_{\mathrm{cl}}$, in an elliptical annulus around the core (from Rc to 2Rc in the case of a circular core), and then subtract this contribution to Σ within the core ellipse, to then yield a mass, ${M}_{c,\min }.$

As discussed by T13, depending on the 3D structure of the core and clump, there can be large differences between ${M}_{c,\max }$ and ${M}_{c,\min }$. In some favorable situations where ${{\rm{\Sigma }}}_{\mathrm{cl}}$ is relatively small, both methods will yield similar mass estimates. However, there can be cases where the surrounding "clump" has a higher inferred mass surface density, and so ${M}_{c,\min }$ is formally negative; i.e., this mass estimate is not well-defined. In fact, this situation arises for C9A, because the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) core is not co-located with a MIREX Σ peak. Another problem with the MIREX-based mass estimates is that the maps can suffer "saturation" in high Σ regions, i.e., where ${\rm{\Sigma }}\gtrsim 0.5\ {\rm{g}}\ {\mathrm{cm}}^{-2}$ (see BT12). This leads to the MIREX-based mass estimate being an underestimate of the true core mass.

As a second method, we calculate core masses from their 1.3 mm continuum emission, ${M}_{c,\mathrm{mm}}$. We use Equation (7) from T13. For consistency, we also assume a dust temperature of ${T}_{d}=10\pm 3$ K (such cold temperatures are expected for highly deuterated cores; see also discussion of T13) and adopt ${\kappa }_{\nu }=5.95\times {10}^{-3}\,{{\mathrm{cm}}^{2}{\rm{g}}}^{-1}$ (Ossenkopf & Henning 1994). We adopt a 30% uncertainty for ${\kappa }_{\nu }$. We do not attempt to carry out clump envelope subtraction with this method, because the ALMA observations tend to filter out the larger scale emission from the clump. Because the uncertainties in temperature cause quite asymmetric uncertainties in ${{\rm{\Sigma }}}_{c,\mathrm{mm}}$ and ${M}_{c,\mathrm{mm}}$, we calculate lower and upper boundaries and put them in sub- and super- scripts in Table 3. Overall, including temperature, opacity, and distance uncertainties, the total uncertainty is about a factor of two in the mass estimation (see detailed discussion in T13); however, because of the difficulties of clump envelope subtraction in the MIREX mass estimates, we prefer the mm continuum-based mass estimate as our fiducial method. We note that, for a "core" contained within the synthesized beam size, at a distance of 5 kpc and with a dust temperature of 10 K, the 1σ mass sensitivity is $0.51\ {M}_{\odot }$.

With these methods, we find that C9A is the most massive of the six cores suitable for dynamical analysis, with ${M}_{c,\mathrm{mm}}\simeq 70\ {M}_{\odot }$ in the Case 1 (i.e., inner) ellipse. The other cores are at least ten times smaller in mass. Considering the Case 2 (i.e., larger) ellipses, C9A rises in mass to ${M}_{c,\mathrm{mm}}\simeq 170\ {M}_{\odot }$. On these scales, the other cores have about $10\ {M}_{\odot }$.

4.5. Core Dynamics

Following T13 and MT03, we consider virialized singular polytropic quasi-spherical cores that have surfaces in approximate pressure equilibrium with their surrounding clump environments. Such cores have a radius ${R}_{{\rm{c}},\mathrm{vir}}$ and internal velocity dispersion ${\sigma }_{{\rm{c}},\mathrm{vir}}$ (assuming some contribution from large scale B-fields such that the Alfvén Mach number in the core is unity). The two quantities are calculated following T13 Equations (2) and (4), given core mass Mc and clump mass surface density ${{\rm{\Sigma }}}_{\mathrm{cl}}$. For each core, we calculate the properties in three cases: (1) using the core mass derived from the MIREX map without envelope subtraction; (2) using the core mass from the MIREX map with envelope subtraction; (3) using the core mass derived from the 1.3 mm continuum map. The clump mass surface density is estimated from the MIREX map in the region from Rc to 2Rc (in the case of a circular core). Table 3 lists the results for the six considered cores.

For our preferred mass estimates via 1.3 mm continuum emission, we find a mean ratio of the observed to predicted virial velocity dispersion for the six cores of $\langle {\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}/{\sigma }_{{\rm{c}},\mathrm{vir},\mathrm{mm}}\rangle =0.80$. This is very similar to the result found by T13 of a ratio of 0.83 for the six cores they analyzed. It suggests that this core population (of 12 cores) has properties that are consistent with the virial equilibrium assumption of the MT03 Turbulent Core model.

The most massive core, C9A, has a ratio of 0.71, which is modestly subvirial. As discussed by T13, apparently subvirial cores may indicate that stronger large scale magnetic fields are present: the fiducial MT03 model assumes that large-scale B-fields are present, which would produce an Alfvén Mach number of unity for turbulence in the core. T13 argued that stronger B-fields, ∼1 mG, were needed in the massive cores of their sample, especially C1-S, if the cores were to be in virial equilibrium. We carry out a similar analysis here, to work out what B-field strengths are needed for virial equilibrium, summarizing the results in Table 4. The magnetic field strengths that are implied by the fiducial MT03 model are again ∼1 mG. They need to be raised modestly to achieve precise virial equilibrium. Also listed is the B-field strength, ${B}_{{\rm{c}},\mathrm{crit}}$, needed to make the core mass equal to the magnetic critical mass (Bertoldi & McKee 1992): these also tend to be slightly higher.

Table 4.  Dynamical Properties of the Six "Best" ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ Cores

Core property (% error) C9A B1A H2A B1B B2A F4A
${R}_{{\rm{c}}}$ (0.01 pc) (20%) 4.71 1.92 2.00 0.822 0.867 1.09
${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ 0.606 ± 0.028 0.395 ± 0.026 0.304 ± 0.029 0.347 ± 0.062 0.255 ± 0.029 0.375 ± 0.049
${{\rm{\Sigma }}}_{\mathrm{cl}}$ $({\rm{g}}\ {\mathrm{cm}}^{-2})$ (30%) 0.317 0.394 0.321 0.361 0.293 0.313
${M}_{{\rm{c}},\mathrm{mm}}$ $({M}_{\odot })$ ${69.7}_{31.7}^{146}$ ${4.91}_{2.24}^{10.3}$ ${4.16}_{1.90}^{8.71}$ ${5.19}_{2.36}^{10.9}$ ${2.26}_{1.03}^{4.73}$ ${2.39}_{1.09}^{5.00}$
${\alpha }_{{\rm{c}}}\equiv 5{\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}^{2}{R}_{{\rm{c}}}/({{GM}}_{{\rm{c}},\mathrm{mm}})$ a ${0.287}_{0.124}^{0.636}$ ${0.702}_{0.299}^{1.56}$ ${0.514}_{0.211}^{1.14}$ ${0.220}_{0.0775}^{0.501}$ ${0.288}_{0.115}^{0.644}$ ${0.742}_{0.288}^{1.66}$
${n}_{{\rm{H}},{\rm{c}},\mathrm{mm}}$ $({10}^{5}\,{\mathrm{cm}}^{-3})$ ${45.9}_{25.5}^{93.8}$ ${48.0}_{26.6}^{98.1}$ ${35.9}_{19.9}^{73.4}$ ${644}_{357}^{1320}$ ${239}_{133}^{489}$ ${128}_{71.0}^{261}$
${t}_{{\rm{c}},\mathrm{ff}}$ $({10}^{5}\,\mathrm{yr})$ b ${0.204}_{0.142}^{0.273}$ ${0.199}_{0.139}^{0.267}$ ${0.230}_{0.161}^{0.309}$ ${0.0544}_{0.0380}^{0.0730}$ ${0.0892}_{0.0624}^{0.120}$ ${0.122}_{0.0853}^{0.164}$
Bc (μG) (mA = 1) ${1220}_{903}^{1750}$ ${812}_{598}^{1160}$ ${541}_{393}^{780}$ ${2620}_{1800}^{3830}$ ${1170}_{842}^{1690}$ ${1260}_{899}^{1830}$
${R}_{{\rm{c}},\mathrm{vir},\mathrm{mm}}$ (0.01 pc) ${11.0}_{16.3}^{7.17}$ ${2.62}_{3.89}^{1.71}$ ${2.67}_{3.97}^{1.74}$ ${2.81}_{4.18}^{1.83}$ ${2.06}_{3.06}^{1.34}$ ${2.05}_{3.05}^{1.34}$
${R}_{c}/{R}_{{\rm{c}},\mathrm{vir},\mathrm{mm}}$ ${0.429}_{0.288}^{0.657}$ ${0.733}_{0.493}^{1.12}$ ${0.749}_{0.504}^{1.15}$ ${0.293}_{0.197}^{0.448}$ ${0.421}_{0.283}^{0.646}$ ${0.531}_{0.357}^{0.814}$
${\sigma }_{{\rm{c}},\mathrm{vir},\mathrm{mm}}$ $(\mathrm{km}\ {{\rm{s}}}^{-1})$ ${0.849}_{0.681}^{1.03}$ ${0.462}_{0.371}^{0.561}$ ${0.421}_{0.338}^{0.511}$ ${0.458}_{0.368}^{0.556}$ ${0.353}_{0.283}^{0.429}$ ${0.364}_{0.292}^{0.442}$
${\sigma }_{{{\rm{N}}}_{2}{{\rm{D}}}^{+}}/{\sigma }_{{\rm{c}},\mathrm{vir},\mathrm{mm}}$ ${0.714}_{0.588}^{0.890}$ ${0.854}_{0.703}^{1.06}$ ${0.723}_{0.595}^{0.901}$ ${0.758}_{0.624}^{0.944}$ ${0.722}_{0.594}^{0.900}$ ${1.03}_{0.850}^{1.29}$
${\phi }_{{\rm{B}},\mathrm{vir}}$ ${6.87}_{3.82}^{11.5}$ ${4.27}_{2.37}^{7.16}$ ${6.65}_{3.70}^{11.2}$ ${5.87}_{3.26}^{9.85}$ ${6.68}_{3.71}^{11.2}$ ${2.58}_{1.43}^{4.32}$
${m}_{{\rm{A}},\mathrm{vir}}$ ${0.519}_{0.383}^{0.772}$ ${0.711}_{0.506}^{1.18}$ ${0.529}_{0.390}^{0.791}$ ${0.573}_{0.419}^{0.874}$ ${0.528}_{0.389}^{0.788}$ ${1.08}_{0.704}^{3.36}$
${B}_{{\rm{c}},\mathrm{vir}}$ (μG) ${2350}_{1360}^{3670}$ ${1140}_{582}^{1830}$ ${1020}_{564}^{1620}$ ${4570}_{2500}^{7230}$ ${2220}_{1200}^{3530}$ ${1160}_{307}^{1980}$
${B}_{{\rm{c}},\mathrm{crit}}$ (μG) ${2830}_{1700}^{4730}$ ${1200}_{725}^{2010}$ ${939}_{565}^{1570}$ ${6930}_{4170}^{11600}$ ${2710}_{1630}^{4530}$ ${1820}_{1090}^{3040}$

Notes.

aVirial parameter (Bertoldi & McKee 1992). bCore free-fall time, ${t}_{{\rm{c}},\mathrm{ff}}={[3\pi /{(32G{\rho }_{c})]}^{1/2}={1.38}^{5}({n}_{{\rm{H}},{\rm{c}},\mathrm{mm}}/{10}^{5}{\mathrm{cm}}^{-3})}^{-1/2}\,\mathrm{yr}$.

Download table as:  ASCIITypeset image

4.6. Notes on Individual Cores

4.6.1. C9A

C9A shows the strongest and most extended ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ structure (Figure 5). As discussed above, it shows complex kinematics, potentially indicative of two merging structures, which have now become connected in their ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ emission. C9A appears to be in a very active, chaotic region (see Figures 5 and 11). We see strong detections of 1.3 mm continuum, especially a large filamentary structure that contains C9A. The mm emission also shows evidence for potential fragmentation or localized heating within the C9A structure. We also see strong detections of DCO+, DCN, ${{\rm{C}}}^{18}{\rm{O}}$, and CH3OH from other sources in the vicinity, but not overlapping with C9A. This core is close to an H ii region IRAS18402-0403, in the direction toward the lower right of the figure (mostly out of the FOV), following the mm continuum filament. Battersby et al. (2010) have studied this area (GLM4 clump) with continuum and molecular line data (not including ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$). Zhang et al. (2009) studied this area (their P2 clump) and found two 1.3 mm continuum cores and an NH3 core. They correspond to the continuum sources to the south of C9A (about 7'' away, see Figures 5 and 11). We checked the Herschel 70 μm image in this area. There are two sources. The larger source corresponds to the H ii region. Next to it is a smaller source, about half that size, which corresponds to the continuum source to the lower-right (Figure 11). C9A is at the boundary of the smaller source.

As discussed above, there is extended, strong SiO(5-4) emission in the surroundings of C9A, likely driven from the strong continuum sources to the lower left. However, C9A also has a weak, small SiO(5-4) counterpart. This only shows up on the red side of the spectrum. The small SiO counterpart is detected from ∼82 to 89 $\mathrm{km}\ {{\rm{s}}}^{-1}$, whereas C9A is detected from 77.05 to 80.35 $\mathrm{km}\ {{\rm{s}}}^{-1}$. No SiO emission on the blue side is detected (the lower limit of velocity is ∼22.5 $\mathrm{km}\ {{\rm{s}}}^{-1}$). We also check that there is no overlap between the ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ core and the SiO connected structures in the PPV space, again using the graph method. It is unclear whether the small SiO(5-4) counterpart in the C9A ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ core is part of the outflow from the external strong continuum source, or whether it indicates the presence of an already-formed protostar within the core (similar to that present in C1-S Feng et al. 2016; Tan et al. 2016). This question will be investigated further by M. Liu et al. (2016, in preparation).

4.6.2. B1A

B1A appears to be in a more quiescent, isolated environment. It shows an elongated morphology, and has relatively weak 1.3 mm continuum counterpart (compared to C9A). We have detected SiO(5-4) emission to the north of B1A outside the continuum boundary. This does not appear to be originating from a protostellar source within the B1A core. However, due to the limited spectral coverage in the blue wing side, we cannot see if there is a symmetric SiO structure on the other side of the B1A core.

Schlingman et al. (2011) have studied this region as one of the BGPS sources (Aguirre et al. 2011 with 33" spatial resolution). They observed HCO+(3-2) and ${{\rm{N}}}_{2}{{\rm{H}}}^{+}$(3-2) lines with 30'' resolution. Their study is on the clump scale. At this scale, they found a velocity dispersion of about 1.5 $\mathrm{km}\ {{\rm{s}}}^{-1}$, which is about a factor of five larger than our measurement at core scale with ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2). We checked the Herschel 70 μm image in this area. B1A has no 70 μm counterpart, and is in a 70 μm dark area.

4.6.3. H2A

H2A appears to be in a quiescent environment. It has no detectable SiO(5-4) emission and a relatively weak 1.3 mm continuum counterpart (compared to C9A). H2A's continuum shows an "X-shaped" structure. We checked the Herschel 70 μm image in this area, finding that H2A has no corresponding 70 μmm source.

4.6.4. B1B

B1B is about 10'' away from B1A (Figure 2). It is displaced in velocity from B1A by about 1 $\mathrm{km}\ {{\rm{s}}}^{-1}$. It has a strong 1.3 mm continuum counterpart (Figure 11). There is a MIR-bright source (visible as a "hole" in the MIREX map) nearby.

4.6.5. B2A

B2A corresponds to a continuum core that is at the end of a group of 1.3 mm continuum sources that also contains a MIR-bright source.

4.6.6. F4A

F4A corresponds to one of a pair of spatially adjacent 1.3 mm continuum sources. The other source does not show significant ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$emission. Overall, this is a relatively quiescent environment.

5. DISCUSSION AND CONCLUSIONS

We have carried out a survey of 32 IRDC clumps, designed to detect cores with strong ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) emission. Such cores may be massive analogs of low-mass pre-stellar or early-stage protostellar cores. This work stems from the pilot study of T13, which identified six such cores in four IRDC clumps. Our current survey has a lower line sensitivity level than T13, but a similar 1.3 mm continuum sensitivity. The spectral set-up includes several ancillary line tracers, such as DCO+(3-2), DCN(3-2), C18O(2-1) and SiO(5-4). We have also utilized the MIREX maps of these regions developed by Butler et al. (2014) and BT12.

In order to process the larger number of target regions, we have presented a new way to automatically identify ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) cores as connected structures in PPV space using Graph theory methods.

In total, 141 ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) core candidates were identified via these automated methods, although many of the weakest sources are likely to be noise fluctuations. The locations of these sources are identified in our maps of the clump-scale regions. We have presented properties of the strongest 50 cores, including their mean velocities and velocity ranges, as well as their ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) line fluxes. We have presented zoom-in maps of the top 15 of these cores, and a dynamical analysis of the best six among these sources.

The main results are: the identification of the very massive (up to $\sim 170\ {M}_{\odot }$) C9A ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) "core," i.e., a connected structure in ppv space, which shows complex structure and kinematics. Several other $\sim 10\ {M}_{\odot }$ cores are found. The ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) velocity dispersions are consistent with the predictions of the turbulent core model of MT03, based on quasi virial equilibrium of such structures. Further follow-up work is needed to test for the starless nature of these cores, especially examining outflow tracers.

The methods presented in this study should also be applied to larger samples of clumps to identify their ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) core populations, which may be key for understanding the origin of the stellar initial mass function and the formation of star clusters.

We thank Brian Svoboda for helpful comments and suggestions. JCT and SK acknowledge an NRAO/SOS grant and NSF grant AST1411527. PC acknowledges the financial support of the European Research Council (ERC; project PALs 320620). This paper makes use of the following ALMA data: ADS/JAO.ALMA#2013.1.00806.S. ALMA is a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada), NSC and ASIAA (Taiwan), and KASI (Republic of Korea), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO, and NAOJ. The National Radio Astronomy Observatory is a facility of the National Science Foundation operated under cooperative agreement by Associated Universities, Inc.

Facility: ALMA - Atacama Large Millimeter Array.

Footnotes

  • We note that the C2 region was studied by Zhang et al. (2015), including an observation of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}\,$ with a sensitivity of 0.0075 $\mathrm{Jy}\ {\mathrm{bm}}^{-1}\ $ per 0.7 $\mathrm{km}\ {{\rm{s}}}^{-1}$. They detected several mm continuum cores in this region, most of which appear to be protostellar.

  • Also studied by Henshaw et al. (2016) with ALMA Band 7 observations, detecting multiple sub-mm continuum cores, including some embedded in very narrow filamentary structures.

  • The choice of a 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$integration range makes this analysis consistent with that of T13, which itself was motivated by our previous single dish studies of these IRDCs. The line-width of ${{\rm{N}}}_{2}{{\rm{D}}}^{+}$(3-2) is ∼1 $\mathrm{km}\ {{\rm{s}}}^{-1}$, so a 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$integration should be enough to include all the cores in these regions, while limiting the amount of extraneous noise. We have carried out tests with 3, 4, 5, and 6 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$integrations, and find comparable results. If a 4σ contour shows up in the 5 $\mathrm{km}\ {{\rm{s}}}^{-1}\,$integration (fiducial), there are signals at the same position in other maps.

  • 10 
Please wait… references are loading.
10.3847/1538-4357/834/2/193