Infrared Dark Clouds (IRDCs)

An IRDC is a dense, cold region of gas and dust in the interstellar medium that appears dark against the bright infrared background of the Milky Way. These clouds are cold (typically around 10-20 K) and dense (gas number densities ~105 cm-3 which is much higher than the average interstellar medium), and contain large amounts of dust that absorb and block background infrared light, causing them to appear as dark silhouettes in infrared images (e.g., from the Spitzer Space Telescope).

IRDC G28
A Spitzer image of the IRDC G28 (aka the Dragon IRDC).
IRDCs are ideal birthplaces of massive stars and star clusters because their high densities and low temperatures allow gravity to overcome internal pressure, leading to gravitational collapse and star formation. In particular, the large mass reservoir in IRDC favors the formation of high-mass stars (more than 8 solar masses) which are rare and short-lived but critically influence galactic evolution. Therefore, observing star-forming activities in IRDCs will improve our understanding of the physical process of massive star formation.

Millimeter wavelengths observations with interferometers:

I have been studying star formation in distant IRDCs using interferometers at mm/sub-mm wavelengths. These wavelengths contain dust continuum emission and molecular line emission that can penetrate the thick gas and dust that obscure the cloud at shorter wavelengths.

G28 analysis
Left: Spitzer image of G28 (background) with star-forming cores (green) and bipolar CO outflows (blue and red). Right: core mass function (top), core growth statistics (middle), and outflow-filament alignment (bottom).

Using the Atacama Large Millimeter/submillimeter Array (ALMA), I have mapped the IRDC G28 in the 1.3 mm dust continuum along with multiple molecular line emissions. In the figure above, the green color represents dust emission, which traces the dense cores deeply embedded within the cloud. These cores are localized regions of high density where individual stars or multiple systems are formed. As shown, the dense cores align well with the dark lane seen in the infrared emission. This alignment reveals that although this IRDC appears quiescent, it is actively forming a cluster of stars.

Star formation activity is further evidenced by CO molecular outflows. In the figure, the blueshifted and redshifted high-velocity CO gas forms bipolar outflows originating from the dense cores, indicating ongoing accretion onto central protostars. These protostars will continue accreting material from their surroundings until they eventually expel the remaining gas, unveiling themselves as a nascent star cluster. The expelled gas will return to the interstellar medium, feeding the next cycle of molecular cloud and star formation.

With the ALMA data, I studied the core mass distribution in G28, showing a shallower core mass function (CMF) than the typical stellar initial mass function (IMF). Similar results were seen in other massive star-forming regions (e.g., the ALMA-IMF Large Program). Either the CMF has little effect on the final IMF, or additional physics is needed to explain the slope change. The results were reported in Kong (2019), ApJ, 873, 31.

By examining the outflows in G28, one can see that they are roughly orthogonal to the filamentary dark lanes. To investigate this further, I conducted a statistical analysis of the relative orientation between the outflows and the filaments, which I presented in Kong et al. (2019), ApJ, 874, 104. The result shows that the relative orientation tends to be close to 90 degree, suggesting the presence of a large-scale physical mechanism that maintains this alignment. Such a mechanism could be closely related to, or even responsible for, the formation of the cloud itself.

Using the CO outflow data, one can distinguish between cores with and without signs of star formation, at least to some extent. Assuming that cores exhibiting outflows are more evolved than those without, comparing the two types can reveal whether core mass increases over time. In Kong et al. (2021), ApJ, 912, 156, I conducted a statistical test of the core masses and found that protostellar cores tend to be more massive than starless ones. This result suggests that dense cores are not static in mass but instead actively interact with their surroundings by continuing to accrete material.

G28 binary
From top-left to lower-left clockwise, the figure gradually zooms in to show the hierarchical structure of a dense core and the embedded binary. The core is at the southern end of G28. See Kong et al. (2023), ApJ, 950, 187 for details.
Using the ALMA C9 array configuration, we observed the massive dense core C1-S1 located at the southern end of G28. This region exhibits the highest extinction (Lim et al. (2016), ApJL, 829, 19) in the cloud, which is why we pay particular attention on it. Given its substantial mass reservoir, C1-S1 is a strong candidate for forming a massive star. The C9 configuration offers ALMA's second-highest spatial resolution in Band 6 (~0.02", or about 100 AU), enabling us to resolve fine structural details. Thanks to ALMA's exceptional sensitivity, we were able to witness the formation of a binary system within the massive C1-S1 core. This discovery aligns with current understanding that massive stars almost always form in binary or multiple systems. Remarkably, our observations capture the initial stages of such a system, providing critical constraints on theories of massive star formation. The results were reported in Kong et al. (2023), ApJ, 950, 187.