Far‐Infrared Spectroscopy of Normal Galaxies: Physical Conditions in the Interstellar Medium

Abstract

The most important cooling lines of the neutral interstellar medium (ISM) lie in the far-infrared (FIR). We present measurements by the Infrared Space Observatory Long Wavelength Spectrometer of seven lines from neutral and ionized ISM of 60 normal, star-forming galaxies. The galaxy sample spans a range in properties such as morphology, FIR colors (indicating dust temperature), and FIR/Blue ratios (indicating star-formation activity and optical depth).In two-thirds of the galaxies in this sample, the [C II] line flux is proportional to FIR dust continuum. The other one-third show a smooth decline in L [CII] /L FIR with increasing F ν (60 µm)/F ν (100 µm) and L FIR /L B , spanning a range of a factor of more than 50. Two galaxies, at the warm and active extreme of the range have L [CII] /L FIR < 2 × 10 −4 (3σ upper limit). This is due to increased positive grain charge in the warmer and more active galaxies, which leads to less efficient heating by photoelectrons from dust grains.The ratio of the two principal photodissociation region (PDR) cooling lines L [OI] /L [CII] shows a tight correlation with F ν (60 µm)/F ν (100 µm), indicating that both gas and dust temperatures increase together. We derive a theoretical scaling between [N II](122 µm) and [C II] from ionized gas and use it to separate [C II] emission from neutral PDRs and ionized gas. Comparison of PDR models of Kaufman et al. (1999) with observed ratios of (a) L [OI] /L [CII] and (L [CII] + L [OI] )/L FIR and (b) L [OI] /L FIR and F ν (60 µm)/F ν (100 µm) yields far-UV flux G 0 and gas density n. The G 0 and n values estimated from the two methods agree to better than a factor of 2 and 1.5 respectively in more than half the sources.The derived G 0 and n correlate with each other, and G 0 increases with n as G 0 ∝ n α , where α ≈ 1.4 . We interpret this correlation as arising from Strömgren sphere scalings if much of the line and continuum luminosity arises near star-forming regions. The high values of PDR surface temperature (270 − 900 K) and pressure (6 × 10 4 − 1.5 × 10 7 K cm −3 ) derived also support the view that a significant part of grain and gas heating in the galaxies occurs very close to star-forming regions. The differences in G 0 and n from galaxy to galaxy may be due to differences in the physical properties of the star-forming clouds. Galaxies with higher G 0 and n have larger and/or denser star-forming clouds.

LIGHTING THE DARK MOLECULAR GAS: H2 AS A DIRECT TRACER

“…The theoretical model of varying a CO by Wolfire et al (2010), assuming H 2 column density of 10 22 cm −2 in a molecular cloud, agrees well with observational results based on dust-mass at low metallicity (Leroy et al 2011;Sandstrom et al 2013). Adopting a solar metallicity value of 8.66 and heating rate/H atom ( ) G n log 0 = −0.3 (from the L L OI FIR studies of Malhotra et al 2001), the Wolfire et al (2010) model also agrees well with our H 2 rotational emission modeled masses.…”

ALMA observations of atomic carbon inz ∼ 4 dusty star-forming galaxies

“…Results were calculated by integrating to A V = 7, and with a cosmic ray flux rate 100 times that of the Milky Way. For reference, the approximate values for Main Sequence galaxies (Malhotra et al 2001), local starbursts (Stacey et al 1991), and local ULIRGs (Davies et al 2003) are shown. The average density for the four DSFGs without a constraint on G 0 is shown at the bottom.…”

Dusty star-forming galaxies at high redshift

“…Graciá-Carpio et al, 2011) at high-redshift have all been detected in [CII]. The line is thought to be an important coolant in the ISM, and can make up as much as 1% of the far infrared luminosity of a galaxy (Nikola et al, 1998;Malhotra et al, 2001;Stacey et al, 2010). However, the sites of origin of [CII] emission can be diverse, and thus its power as a diagnostic for the physical conditions in the ISM of early Universe galaxies is still debated.…”

“…However, present-epoch ULIRGs emit roughly ∼ 10% of the expected [CII] flux, given their far infrared luminosities (e.g. Malhotra et al, 1997Malhotra et al, , 2001Luhman et al, 1998Luhman et al, , 2003. A variety of possible explanations have been posited for the apparent deficit.…”