Abstract

Molecular hydrogen in the interstellar medium of galaxies

Naslim Neelamkodan (NAOJ)

Ciska Kemper (ASIAA) Masaaki Otsuka (ASIAA) Oscar Morata (ASIAA) Sundar Srinivasan (ASIAA) Akiko Kawamura (NAOJ) Suzanne Madden (CEA/Saclay) Sacha Hony (Heidelberg)

Tracing molecular gas in metal-poor environment of nearby galaxies is crucial for understanding the process of star formation in the early universe. Unfortunately, the most abundant molecule, H2, remains largely undetectable due to its symmetric structure and the lack of a permanent dipole moment. Therefore, to quantifying the molecular gas reservoir, we usually rely on the second most abundant molecule, CO, using a CO-to-H2 conversion factor. However, CO can have difficulties in tracing all of the molecular gas under certain conditions, such as low-metallicity environment when substantial H2 may exist outside of CO emitting regions (Remy-Ruyer et al. 2014; Chevance et al. 2016). In a metal-poor ISM, due to the reduced dust-to-gas ratio, most of the CO can be photo-dissociated by hard UV radiation field, however H2 is self-shielded from the UV radiation field and expect to reside everywhere in the ISM. The pure rotational 0–0 transitions of H2 due to the molecule’s quadrupole moment are more direct tracers of the H2 gas, although they are only excited at higher temperatures than those prevalent in molecular cloud interiors. These mid-infrared transitions trace the bulk of the warm molecular gas with temperatures between 100 and 1000 K, which is a small but non-negligible fraction of the total molecular gas reservoir. The infrared spectrograph on the Spitzer space telescope enabled us to detect these weak H2 transitions in many nearby low-metallicity galaxies including our neighbors, the Magellanic Clouds, and establish that around 10% of the gas was in the warm molecular phase (Naslim et al. 2015). However, The uncertainty of the amount of H2, physical properties and excitation conditions in galaxies are exacerbated due to the relatively weak detection threshold of H2.

Our earlier work (Naslim et al. 2015) was based on 1’ IRS spectral maps obtained with Spitzer. The pure rotational 0-0 transitions of H2 at 28.2 (S(0)) and 17.1 μm (S(1)) are detected, however the higher level transitions are almost exclusively upper-limit measurements. The single-temperature fits through the lower transition lines give temperatures in the range 86–137 K. The bulk of the excited H2 gas is found at these temperatures and contributes ∼5–17 percent to the total gas mass. We found a tight correlation of the H2 surface brightness with polycyclic aromatic hydrocarbon and total infrared emission, which is a clear indication of photoelectric heating in photo-dissociation regions. The excitation of H2 by this process is equally efficient in both atomic- and molecular-dominated region and we cannot rule-out the possibility of shock excitation of H2 in our samples. The H2 transitions are found to be important diagnostic tool for shocks in many galactic and extragalactic environments. The excitation diagram of H2 is generally characterized by 1) the excitation temperature of the thermal component 2) the ortho/para ratio of non-thermal component. In order to clearly calibrate the ortho/para ratio and for a satisfactory explanation of excitation mechanism we require the detection of H2 higher transitions at higher sensitivity. This will be possible with the unique capability of JWST. The mid-IR spectrograph (MIRI) on JWST covering a wavelength range 4.6-28.6 μm will be able to detect and characterize H2 0-0 rotational transitions in nearby metal-poor galaxies at a higher sensitivity and spectral resolution. MIRI's integral field unit will allow us to investigate the physical state and kinematics of warm molecular gas that is dynamically heated by star formation or shocks. We plan to probe H2 emission in galaxies of different metallicities and study their physical conditions and excitation mechanisms to investigate any trend with metallicity.

Talk