We discuss results from numerical simulations of star formation under various environmental conditions ranging from the turbulent interstellar medium to low-mass halos in the early universe. The thermodynamic behavior of the star-forming gas plays a crucial role in fragmentation and determines the stellar mass function as well as the dynamic properties of the nascent stellar cluster. The thermodynamic state of the gas is a result of the balance between heating and cooling processes, which in turn are determined by atomic and molecular physics and by chemical abundances. Features in the effective equation of state of the gas, such as a transition from a cooling to a heating regime, define a characteristic mass scale for fragmentation and so set the peak of the initial mass function of stars (IMF). As it is based on fundamental physical quantities and constants, this is an attractive approach to explain the apparent universality of the IMF in the solar neighborhood as well as the transition from purely primordial high-mass star formation to the low-mass mode observed today.

We have used the new broadband capabilities of the Mopra telescope to map the distribution of 26 different molecular transitions in an approximately 1 degree square region of the southern Galactic plane (the G333/RCW106 giant molecular cloud complex). The aim is to addresss observationally some of the key questions about the dynamical processes surrounding massive star formation (e.g. massive stellar winds and large-scale galactic flows) and their relative importance in regulating the star formation process. These dynamical processes help drive the turbulent motions, which are ubiquitous in giant molecular clouds (GMCs). The multi-molecular line nature of this survey is what distinguishes it from similar surveys and is crucial for gaining a clear picture of the energetics and dynamics of the gas. Investigating and understanding the chemistry of this region is a necessary part of the project if the molecular line observations are to be interpreted physically.

We present 3D high resolution numerical simulations of turbulent converging interstellar atomic hydrogen flows. First, we report pure hydrodynamical simulations from which we infer statistics of the CNM structures. We find that the CNM structures are characterized by a mass spectrum dN/dM ∝ M-1.7-1.8, a mass-size relation M ∝ L2.3-2.5, and a velocity dispersion σ ∝ L0.4-05. These are reminiscent of what has been inferred for CO molecular clumps. Then we present a 3D calculation, which includes magnetic field and self-gravity with the purpose of studying the formation of molecular clouds.

Observations of the interstellar medium (ISM) reveal its fractal structure over a large range of scales. The quantitative description shows that the ISM consists mainly of surfaces, so that the stellar UV-radiation can deeply penetrate into the ISM and in consequence dominates its physical and chemical conditions. Thus, the bulk of the ISM can be identified as photon dominated regions (PDRs). A simple model with the proper fractal characteristics is that of an ensemble of spherical clumps with a power-law mass spectrum and a power-law mass-size relation. Hence, we model the FIR line emission of the molecular gas by the integrated emission of such an ensemble, calculating the emission of individual clumps with the KOSMA-τ PDR code. This clumpy PDR-model, with observationally constrained parameters, is fitted to the CO, [C i], and [O i] emission along the Galactic plane as observed by the FIRAS instrument abord the COBE satellite. The FIR line intensity distribution with Galactic longitude is reproduced within factors of about 2. The predicted [C ii] 158 μm line intensity and FIR continuum emission also coincide with the observations.

The NANTEN2-4 m submillimeter telescope was used to map the [C i] 3P1-3PP0, 3P2-3P1 and CO 4–3, 7–6 lines in two 4'×4' regions of Carina where molecular material interfaces with radiation from the massive star clusters. These data were combined with 13CO SEST spectra, HIRES/IRAS maps at 60 µm, 100 µmm, maps of 8 µmm IRAC/Spitzer, and of MSX emission. We use PDR models to interpret the observed far-infrared and submillimeter emission, assuming that all observed emission stems from an ensemble of spherically symmetric PDR clumps.

Some recent results on radiative transfer in standard PDR models (1D, stationary) are presented and illustrated with the Meudon PDR code. It is shown that current modeling capacities are sufficient to go beyond the usual (mostly escape probabilities) approximations. A few more sophisticated works adapted to more complex physical conditions are mentioned.

Photodissociation region (PDR) models are used to understand the evolution of the FUV illuminated matter both in our Galaxy and in external galaxies. To prepare for the unprecedented spatial and spectroscopic capabilities of ALMA and Herschel, two different kinds of progresses are currently taking place in the field. First, numerical models describing the chemistry of a molecular cloud are being benchmarked against each other to ensure that all models agree not only qualitatively but also quantitatively on at least simple cases (Röllig et al. 2007). Second, new or improved chemical rates are being calculated/measured by several theoretical and experimental groups. However, the difficulty of this last effort implies that only a few reactions (among the thousands used in chemical networks) can be thoroughly studied. New numerical tools are thus being developed for taking into account the impact of the uncertainties of the chemical rates on the chemical model predictions and their comparison with observed abundances (e.g. Wakelam et al. 2005, 2006). In view of the intrinsic complexity of building reliable chemical networks and models, there is an obvious need for well-defined observations that can serve as basic references. PDRs are particularly well suited to serve as references because they are the link between diffuse and dark clouds, thus enabling to probe a large variety of physical and chemical processes. We describe the merit of two prototypical PDRs: the Orion Bar and the Horsehead mane.

We have expanded our model of photodissociation regions (PDRs) to include the freezing of O- and C-bearing species on dust grains, simple grain surface chemistry, and desorption processes, including photodesorption, that may be important in the surface layers of diffuse, translucent, and dense molecular clouds. The main result of including these processes is that a number of important gas-phase species, including H2O and O2, peak in abundance at AV ~ few into the cloud. Most of the gas-phase column, and most of the emission, from these species arises in the peak. Closer to the surface, H2O and O2 are photodissociated, while deeper into the cloud, they freeze out onto grain surfaces. The result is H2O and O2 column densities that are nearly constant for a wide range of gas densities, n, and for FUV fields G0$\lesssim$ 500. The roughly constant column densities of these species provides an explanation for the low line-of-sight average abundances of H2O observed toward GMCs. The model results also suggest that regions of high FUV field are the best places to search for O2.

The nuclei of active galaxies harbor massive young stars, an accreting central black hole, or both. Results are presented for molecular gas that is exposed to X-rays (1–100 keV, XDRs) and far-ultraviolet radiation (6–13.6 eV, PDRs). Particular attention is paid to species like HCN, HNC and HCO+. Line ratios of HCN/HCO+ and HNC/HCN discriminate between PDRs and XDRs.

I review recent studies of the emission-line regions in Orion and M 17. Both have similar geometries, a bubble of hot shocked gas surrounding the central star cluster, with H+, H0, and H2 regions, often referred to as H II regions, PDRs, and molecular clouds, forming successive shells on the surface of a molecular cloud. The magnetic fields in the H0 regions have been measured with 21 cm Zeeman polarization and are found to be 1 – 2 dex stronger than the field in the diffuse ISM. The regions appear to be in rough hydrostatic equilibrium. The H+ region is pushed away from the star cluster by starlight radiation pressure. Since most starlight is in ionizing radiation, most of its outward push will act on the H+ region and then on to the H0 region. The magnetic pressure in the H0 region balances the momentum in starlight and together they set the location of the H0 region. The picture is that, when the star cluster formed, it created a bubble of ionized gas which expanded and compressed surrounding H0 and H2 regions. The magnetic field was amplified until its pressure was able to support the momentum in starlight. This offers a great simplification in understanding the underlying physics that establishes parameters for PDR models.

In this article we review some of the most recent advances in the modelling and interpretation of molecular line emissions from Galactic and extragalctic environments, in particular from massive star forming regions. Hence We show that observations of extragalactic molecules can potentially be used to infer and disentangle the physical conditions in the gas, as it is already routinely done in our own Galaxy.


We report our recent progress on extragalactic spectroscopic and continuum observations, including HCN(J=1–0), HCO+(J=1–0), and CN(N=1–0) imaging surveys of local Seyfert and starburst galaxies using the Nobeyama Millimeter Array, high-J CO observations (J=3–2 observations using the Atacama Submillimeter Telescope Experiment (ASTE) and J=2–1 observations with the Submillimeter Array) of galaxies, and λ 1.1 mm continuum observations of high-z violent starburst galaxies using the bolometer camera AzTEC mounted on ASTE.

We summarize the main highlights from ISO observations towards Sgr B2 and Orion KL in the far–IR domain (~43 to 197 μm). Both Star–Forming Regions are among the best sources to construct a template for more distant and unresolved regions (e.g., extragalactic). We stress some peculiarities in the interpretation (excitation and radiative transfer) of far–IR spectral lines and dust continuum emission.

We are studying the Central Molecular Zone (CMZ) in the inner few degrees around the Galactic Centre, by mapping a large number of 3-mm molecular lines. We have used the 22-m Mopra telescope, with the new capabilities of the 8-GHz bandwidth MOPS digital filter bank. During 2006, we covered a 5 × 5 arcmin2 area of the Sagittarius B2 molecular cloud complex, with spectral coverage of nearly all of the range 81.7 to 113.5 GHz with 4 tunings of the broad band mode. We imaged over 50 spectral lines with strong extended emission and around 110 more lines which were weak, or concentrated at the Sgr B2(N) and Sgr B2(M) cores (Jones et al.  2008). We find substantial differences in chemical and physical conditions within the complex. During 2007 we covered the larger region of longitude -0.2 to 0.9 deg. and latitude -0.20 to 0.12 deg., including Sgr A and Sgr B2, in the frequency range 85.3 to 91.3 GHz. This includes lines of C3H2, CH3CCH, HOCO+, SO, H13CN, H13CO+, SO, HN13C, C2H, HNCO, HCN, HCO+, HNC, HC3N, 13CS and N2H+. 


We present a brief discussion of the statistical surveys of HCN, HNC, HCO+ and HC3N that are used to model the extreme environments in the nuclei of starbursts and AGNs. Molecular studies are particularly useful for probing the deeply enshrouded dusty nuclei of luminous infrared galaxies. Here we present NGC 4418 as an example, one of the closest LIRG with high obscuration of the inner region. The interpretation of the observed line ratios require parallel development of theoretical chemical and radiative transport models.

The use of specific tracers of the dense molecular gas phase can help to explore the feedback of activity on the interstellar medium (ISM) in galaxies. This information is a key to any quantitative assessment of the efficiency of the star formation process in galaxies. We present the results of a survey devoted to probe the feedback of activity through the study of the excitation and chemistry of the dense molecular gas in a sample of local universe starbursts and active galactic nuclei (AGNs). Our sample includes also 17 luminous and ultraluminous infrared galaxies (LIRGs and ULIRGs). From the analysis of the LIRGs/ULIRGs subsample, published in Graciá-Carpio et al.  (2007), we find the first clear observational evidence that the star formation efficiency of the dense gas, measured by the LFIR/LHCN ratio, is significantly higher in LIRGs and ULIRGs than in normal galaxies. Mounting evidence of overabundant HCN in active environments would even reinforce the reported trend, pointing to a significant turn upward in the Kennicutt-Schmidt law around LFIR = 1011L๏. This result has major implications for the use of HCN as a tracer of the dense gas in local and high-redshift luminous infrared galaxies.

Multiline 12CO and 13CO measurements of a large sample of galaxy centers reveal the physical conditions of the pronounced central molecular gas concentrations in spiral galaxies. The molecular gas occurs in at least two components: hot and relatively tenuous, dense and relatively cool. Most of the gas mass is in the former component. Molecular gas is strongly concentrated within in a few hundred parsec from the nucleus, but the central few dozen parsecs are frequently gas-poor. The CO-to-H2 conversion factor is three to ten times lower in galaxy centers than in the Milky Way disk.

We present the results of systematic millimeter interferometric observations of luminous infrared galaxies (LIRGs), using the Nobeyama Millimeter Array. We observed HCN (J = 1–0) and HCO+ (J = 1–0) emission lines simultaneously, and derived their brightness-temperature ratios, to investigate whether the observed ratios are similar to those found in AGN-dominated nuclei or starburst galaxies. LIRGs with (without) luminous buried AGN signatures in our infrared spectra tend to show high (low) HCN (J = 1–0) to HCO+ (J = 1–0) brightness-temperature ratios as seen in galaxy nuclei dominated by AGNs (starbursts). The high ratios in buried AGN candidates could be explained by the enhanced HCN abundance, and/or infrared radiative pumping of the HCN molecule, in the close vicinity of an AGN.

Molecular gas in external galaxies is a subject of crucial importance for observational and theoretical studies of galaxy formation. Compact molecular gas around an active galactic nucleus (AGN) is expected to be an important part of the energy budget of the AGN and/or the nuclear starburst. Recent observational studies suggest that line ratios in the millimeter and submillimeter bands may be a good tool to reveal the long-standing question on the origin of activity – AGN or nuclear starburst. We have constructed a powerful “telescope” of theory, a three-dimensional non-LTE line transfer code, preceding the high resolution and sensitivity observations such as ALMA.