12mm line survey of the dense molecular gas towards the W28 field TeV gamma-ray sources

We present 12mm Mopra observations of dense molecular gas towards the W28 supernova remnant (SNR) field. The focus is on the dense molecular gas towards the TeV gamma-ray sources detected by the H.E.S.S. telescopes, which likely trace the cosmic-rays from W28 and possibly other sources in the region. Using the NH3 inversion transitions we reveal several dense cores inside the molecular clouds, the majority of which coincide with high-mass star formation and HII regions, including the energetic ultra-compact HII region G5.89-0.39. A key exception to this is the cloud north east of W28, which is well-known to be disrupted as evidenced by clusters of 1720MHz OH masers and broad CO line emission. Here we detect broad NH3, up to the (9,9) transition, with linewidths up to 16 km/s. This broad NH3 emission spatially matches well with the TeV source HESS J1801-233 and CO emission, and its velocity dispersion distribution suggests external disruption from the W28 SNR direction. Other lines are detected, such as HC3N and HC5N, H2O masers, and many radio recombination lines, all of which are primarily found towards the southern high-mass star formation regions. These observations provide a new view onto the internal structures and dynamics of the dense molecular gas towards the W28 SNR field, and in tandem with future higher resolution TeV gamma-ray observations will offer the chance to probe the transport of cosmic-rays into molecular clouds.

💡 Research Summary

This paper presents a comprehensive 12 mm molecular line survey of the dense interstellar medium surrounding the W28 supernova remnant (SNR) using the Mopra radio telescope. The authors focus on the regions that coincide with the TeV γ‑ray sources detected by the H.E.S.S. array, aiming to link the distribution and physical state of the dense gas with the propagation of cosmic‑ray (CR) particles accelerated by the SNR (and possibly other nearby accelerators).

The survey covers roughly five square degrees and simultaneously records nine inversion transitions of ammonia (NH₃) from (1,1) up to the high‑energy (9,9) line, together with ancillary species such as HC₃N, HC₅N, water masers, and a suite of radio recombination lines (RRLs). NH₃ is an excellent tracer of gas with densities of 10⁴–10⁵ cm⁻³; the ladder of inversion transitions provides a direct thermometer and a probe of optical depth, allowing the authors to map temperature, column density, and kinematics across the field.

Two distinct environments emerge from the data. In the north‑eastern molecular cloud, well known for its interaction with the W28 shock, the authors detect very broad NH₃ emission (linewidths up to 16 km s⁻¹) extending to the (9,9) transition. This broad component aligns spatially with clusters of 1720 MHz OH masers, with CO line wings previously reported, and with the TeV source HESS J1801‑233. The velocity dispersion shows a clear gradient pointing toward the SNR, indicating that the shock front is impinging on the cloud, heating it to 40–60 K (significantly above the ∼15 K typical of quiescent clouds) and stirring the gas on sub‑parsec scales. The detection of high‑J NH₃, together with HC₃N and HC₅N, points to shock‑driven chemistry that enhances complex carbon‑chain molecules.

Conversely, the southern part of the field is dominated by high‑mass star‑forming complexes, most notably the ultra‑compact H II region G5.89‑0.39. Here NH₃ emission is confined to the lower‑excitation (1,1)–(3,3) lines, with narrow linewidths (2–4 km s⁻¹) and temperatures around 20–30 K. The same locations host strong water masers, abundant HC₃N/HC₅N, and multiple RRLs, confirming the presence of dense, ionised gas associated with massive young stellar objects. The TeV γ‑ray flux in this region is comparatively low, suggesting that local stellar activity does not dominate the high‑energy emission.

By comparing the dense‑gas maps with the H.E.S.S. γ‑ray images, the authors find a striking spatial coincidence between the broad‑NH₃ component in the north‑east cloud and the peak of the TeV emission. This supports the “crushed‑cloud” scenario in which CR protons accelerated at the SNR shock diffuse into the adjacent molecular material, interact with the dense gas, and produce π⁰‑decay γ‑rays. The lack of a similar correlation in the southern star‑forming zones reinforces the interpretation that the dominant CR population originates from the SNR rather than from local stellar accelerators.

The paper also presents quantitative analyses: rotational temperature diagrams for NH₃, non‑LTE radiative transfer modelling to estimate column densities, and a kinematic decomposition that reveals systematic velocity shifts consistent with shock propagation. The authors discuss the implications for CR diffusion coefficients, suggesting that the dense clumps act as partial barriers that slow CR penetration, thereby shaping the observed γ‑ray morphology.

In summary, this work demonstrates that high‑resolution 12 mm molecular spectroscopy provides a powerful diagnostic of SNR–cloud interactions and of the underlying CR physics. The authors argue that future observations with the Cherenkov Telescope Array (CTA), combined with even higher‑resolution interferometric molecular line studies (e.g., ALMA), will enable direct measurements of CR diffusion lengths, energy‑dependent propagation, and the feedback of CRs on cloud chemistry and dynamics.