Free Access
Issue
A&A
Volume 541, May 2012
Article Number A4
Number of page(s) 30
Section Extragalactic astronomy
DOI https://doi.org/10.1051/0004-6361/201118029
Published online 18 April 2012

© ESO, 2012

1. Introduction

From the first CO observations of IRAS galaxies (Young et al. 1984), much effort has been devoted to understanding spectroscopic observations of molecules in (ultra)luminous infrared galaxies ((U)LIRGs), and how they are related to the conditions traced by the continuum and line observations of atoms and ions. The launch of the Infrared Space Observatory (ISO) and, more recently, the Herschel Space Observatory, have opened a new window to study extragalactic sources in the far-infrared (hereafter, far-IR) domain, where the bulk of the (U)LIRG luminosity is emitted. The relevance of far-IR molecular spectroscopy of (U)LIRGs is well illustrated with the initial ISO result of Fischer et al. (1999), that the fine-structure lines of atoms and ions decrease in strength, as absorption in molecular lines, specifically H2O and OH, increase in depth.

Water (H2O) and hydroxyl (OH) are indeed key molecular species that trace relevant physical and chemical properties of the interstellar medium in luminous infrared extragalactic sources. Powerful OH mega-masers are common in ULIRGs, and H2O masers are found in AGN accretion disks and in post-shocked gas associated with nuclear jets (see Lo 2005, for a review). With many transitions lying at far-IR wavelengths, where the bulk of the luminosity is emitted, H2O and OH are mainly excited through absorption of far-IR photons (González-Alfonso et al. 2004, hereafter G-A04), and thus their excitation helps probe the general properties of the underlying far-IR continuum sources (González-Alfonso et al. 2008). Chemically, H2O is expected to trace an undepleted chemistry where grain mantles are evaporated; both H2O and OH are also boosted in shocks and, especially OH, in X-ray and cosmic ray dominated regions (XDRs and CRDRs). Furthermore, OH has turned out to be a unique tracer of massive molecular outflows (Fischer et al. 2010; Sturm et al. 2011). Recently, H2O has been detected in high-z sources (Impellizzeri et al. 2008; Omont et al. 2011; Lis et al. 2011; van der Werf et al. 2011; Bradford et al. 2011), and thus understanding its role in the local Universe will be crucial for future routine observations of H2O with ALMA in the far Universe. HCN is another key molecule widely studied in local (U)LIRGs at (sub)millimeter wavelengths (e.g., Aalto et al. 1995). NH3, first detected in extragalactic sources by Martin & Ho (1979), is another important tracer; in Arp 220, it was detected at far-IR wavelengths with ISO (G-A04) and at centimeter wavelengths by Takano et al. (2005) and recently by Ott et al. (2011). With its high sensitivity and spectral resolution at far-IR wavelengths, the Herschel/PACS instrument (Pilbratt et al. 2010; Poglitsch et al. 2010) provides new opportunities for using the far-IR transitions of these molecules to probe the kinematic, chemical, and radiative activity in IR luminous galaxies.

We report in this work Herschel/PACS spectroscopic observations of H2O, OH, their 18O isotopologues, NH3, and the surprising detection of highly excited HCN, in NGC 4418 and Arp 220. NGC 4418 is a peculiar, single nucleus galaxy, with a moderate luminosity (~1011 L) but with other properties similar to warm ULIRGs: a high LFIR/MH2 ratio and an extreme [C ii] deficit (e.g. Graciá-Carpio et al. 2011), a high continuum surface brightness, an extremely compact luminosity source (Evans et al. 2003), and warm infrared colors (S25   μm/S60   μm = 0.23, S60   μm/S100   μm = 1.37). We adopt a distance to the source of 29 Mpc. Arp 220 is the prototypical and nearest ULIRG, showing a double nucleus and a luminosity of  ≈1.3 × 1012 L. We adopt a distance to the source of 72 Mpc. The observations are described in Sect. 2; models are shown in Sect. 3; model results are discussed in Sect. 4, and our main conclusions are summarized in Sect. 5.

2. Observations and results

2.1. Observations

The full (52.3−98, 104.6−196 μm), high resolution, PACS spectra of NGC 4418 and Arp 220, taken as part of the guaranteed-time key programme SHINING, were observed on July 27th and February 27th (2010), respectively. The spectra were taken in high spectral sampling density mode using first and second orders of the grating. The velocity resolution of PACS in first order ranges from  ≈320 kms-1 at 105 μm to  ≈180 kms-1 at 190 μm, and in second order from  ≈210 kms-1 at 52 μm to  ≈110 kms-1 at 98 microns. For NGC 4418, we also present a scan around the [O i] 63 μm line taken in third order with velocity resolution of  ≈85 kms-1. The data reduction was mostly done using the standard PACS reduction and calibration pipeline (ipipe) included in HIPE 5.0 975. The reduction of Arp 220 required, however, two additional steps in order to correct for pointing errors that resulted in significant variations of the continuum level of the central spaxel1 as well as in wavelength offsets generated by pointing errors along the dispersion direction of the slit (Poglitsch et al. 2010). These effects were produced by the movement of the source between the central spaxel and a neighboring spaxel. In order to correct for them, (i) the emission from the two spaxels with highest signal were added together in an intermediate step in the pipeline, which significantly improved the continuum baseline of the combined spectrum but increased the noise by a factor2; (ii) the time evolution of the pointing shifts was reconstructed from the star tracker frames, and wavelength calibration was corrected according to Fig. 8 by Poglitsch et al. (2010). Pointing errors across the slit were found from 0 to  ~3′′, resulting in velocity shifts of  ≲60 kms-1. We estimate that the associated uncertainties are below 30 kms-1, though they may be higher for some individual lines.

For the final calibration, the spectrum was normalized to the telescope flux and recalibrated with a reference telescope spectrum obtained from observations of Neptune. The flux density of the spectrum in the central spaxel(s) was then rescaled to the value of the total continuum of the source obtained by the combination of the 25 spaxels of the PACS spectrometer. This method assumes that the distribution of the IR continuum and the lines are similar for the spatial resolution of the observations. Indeed the line to continuum ratio in the central and the combined 25 spaxels was found to be the same for all strong lines except for the ground-state OH 119 and 79 μm in Arp 220, which display 9% and 3% deeper absorption in the central spaxels. The latter OH lines thus appear to show low-level reemission (probably through scattering) over kpc spatial scales. In general, the comparison indicates that both sources are at far-IR wavelengths point-sources compared with the PACS 9′′ × 9′′ spatial resolution (~1.3 × 1.3 kpc2 for NGC 4418 and  ~3.1 × 3.1 kpc2 for Arp 220).

thumbnail Fig. 1

Spectral energy distribution of NGC 4418 and Arp 220 from mid-IR to millimeter wavelengths. Both the Herschel/PACS and the Spitzer/IRS spectra are shown. Data points at (sub)millimeter wavelengths for NGC 4418 are from (Yang & Phillips 2007, 350 μm), (Roche & Chandler 1993, 450, 800, and 1100 μm), (Dunne et al. 2000; Lisenfeld et al. 2000, 850 μm), and (Albrecht et al. 2007, 1300 μm). For Arp 220, they are taken from (Eales et al. 1989, 450 μm); (Rigopoulou et al. 1996, 350, 800, and 1100 μm); and (Sakamoto et al. 1999, 2008, 1300 μm). Models discussed in Sect. 3 are included. For NGC 4418, the yellow, blue, light-blue, and solid-green curves show the models for the hot, core, warm, and extended components, respectively, with parameters given in Table 1. An additional cold component, shown with the dashed-green curve, is included to better fit the SED at long wavelengths. For Arp 220, the yellow, blue, light-blue, and solid-green curves show the models for the hot, western nucleus, eastern nucleus, and extended components, respectively (Table 1).

Open with DEXTER

The continua of both sources are shown in Fig. 1, including the Spitzer IRS spectra (Spoon et al. 2007); the models are discussed in Sect. 3. Both galaxies show strong silicate absorption at 9.7 μm, with PAH emission that is weak in Arp 220 and only weakly detected at 11.2 μm in NGC 4418 (Spoon et al. 2007). Besides the obvious differences in luminosity, a striking contrast between both sources is the S60   μm/S25   μm ratio, which is  ≈4 in NGC 4418 and approaches  ≈13 in Arp 220. The higher value in Arp 220 indicates that larger amounts of dust surround the nuclei in an extended region (ER in G-A04, denoted here as Cextended), both extinguishing the nuclear emission at 25 μm and contributing to the far-IR continuum. This emission can be interpreted as heating of dust by the radiation emanating from the nuclei (Soifer et al. 1999, G-A04). In NGC 4418, the warmer spectral energy distribution (SED) suggests that the amount of dust in Cextended is lower, but still probably significant (see Sect. 3). On the other hand, the (sub)millimeter emission in Arp 220 is dominated by the nuclei as revealed by interferometric (sub)millimeter observations (Downes & Eckart 2007; Sakamoto et al. 2008), indicating extreme opacities in, and high luminosities from the nuclei. No interferometric observations are available for NGC 4418.

Spectroscopic data used for both line identification and radiative transfer modeling were taken from the JPL (Pickett et al. 1998) and CDMS (Müller et al. 2001, 2005) catalogues. Figure 2 shows the energy level diagrams of OH and H2O. The spectra around the H2O lines are shown in Fig. 3 (blue and green bands, λ < 100 μm) and Fig. 4 (red band, λ > 100 μm), while high-lying H2O lines tentatively detected in NGC 4418 (blue-dashed arrows in Fig. 2) are shown in Fig. 5. The OH line profiles are presented in Fig. 6. Figures 7 and 8 show excerpts of the PACS spectra around the O and 18OH lines, respectively. The HCN and NH3 lines are displayed in Figs. 9 and 10, respectively. Figure 11 compares the line profiles of the [O i] 63 and 145 μm transitions with those of some OH lines. Figure 12 compares the line shapes of two HCO+ transitions observed with high spatial resolution by Sakamoto et al. (2009) in Arp 220, with selected lines observed with Herschel/PACS. Results of the continuum and line modeling are summarized in Tables 1 and 2, and the line equivalent widths of the H2O, [O i] and OH, HCN, NH3, O, and 18OH lines are listed in Tables 3 to 8.

2.2. Systemic redshifts

Line parameters were derived by subtracting baselines and fitting Gaussian curves to all spectral features. One relevant question in our study is the zero-velocity adopted for the high-excitation molecular gas in the nuclear region of the galaxies. In NGC 4418, the excited H2O lines do not show any evidence of systematic shifts indicative of outflowing or inflowing gas. Figure 13 indicates that the only H2O line clearly redshifted relative to all others is the ground-state 212 → 101 at 179.5 μm, which is expected to be contaminated by CH+ (2–1) (see below). From all other H2O lines, the average redshift is z = 0.007035 ± 0.000070 (cz = (2110 ± 20) kms-1), where the error denotes the standard deviation of the mean. The inferred redshift is fully consistent with observations of CO, CN, HNC, and other molecular species observed at millimeter wavelengths (e.g. Aalto et al. 2007a). The Sloan Digital Sky Survey (SDSS) spectrum of NGC 4418 also gives a consistent redshift for the optical absorption/emission lines of z = 0.0070 and 0.0072, respectively (D. Rupke, private communication). Finally, this is also consistent with the redshift measured for the high-lying OH lines (see below), which also trace the nuclear component, and with that of the [C ii] emission line at 157.741 μm, z = 0.007050, which we adopt as the reference value. In Arp 220, Sakamoto et al. (2009) detected P-Cygni profiles in excited submillimeter lines of HCO+ toward both nuclei, indicating dense outflowing gas in the nuclei (see also Fig. 12 and Sect. 2.7). The transition from absorption to emission occurs at a velocity of  ≈5420 kms-1 toward both nuclei, which we adopt as the systemic redshift (z = 0.0181). The foreground absorption traced by CO (3–2) toward the western nucleus lies at a lower velocity,  ≈5350 kms-1. The line velocity shifts (Vshift) relative to the above systemic redshifts, the line widths, and the equivalent widths (Weq) are plotted in Fig. 13 for the H2O lines and in Fig. 14 for the OH lines.

thumbnail Fig. 2

Energy level diagrams of OH and H2O (ortho and para). Red arrows indicate lines detected in both sources; blue arrows denote lines detected only in NGC 4418, green-dashed arrows mark transitions contaminated (blended) by lines of other species, and blue-dashed arrows denote marginal detections in NGC 4418. The small numbers in the OH diagram indicate rounded wavelengths in μm.

Open with DEXTER

2.3. H2O and OH

Based on the H2O spectra and the molecular line positions shown in Figs. 3 and 4, most H2O lines are free of contamination from lines of other species, with the following exceptions: as mentioned above, the H2O 212 → 101 line in Fig. 4q at 179.5 μm is blended with CH+; the 808 → 717 line in Fig. 3f is blended with highly excited NH3 (see Fig. 10a); the 524 → 413 (Fig. 3i) and the 221 → 212 lines (Fig. 4q) may have some contribution from CH; finally, the spectral feature at 104 μm associated in Fig. 4a with the 615 → 606 line is probably contaminated by OH+. Other contaminated lines marked with dashed-green arrows in Fig. 2 are not shown. The OH lines in Fig. 6 have relatively weak contamination by NH3 in the red Λ-component of the 84 μm doublet (panel e, see Fig. 10c), and by H2O in the blue Λ-component of the 65 μm doublet (panel f).

The detections summarized in Fig. 2 indicate extreme excitation in the nuclei of both galaxies. In Arp 220, lines with lower level energy up to Elower ≈ 600 K in both H2O (707 → 616 and 717 → 606 , Fig. 3j, the latter contaminated by excited NH3) and OH (Π1/29/2 → 7/2, Fig. 6i) are detected. NGC 4418 shows even higher-lying lines, with detected H2O lines as high as 818 → 707 and 716 → 625 (Figs. 3f, g; Elower ≳ 800 K). These are the highest-lying H2O lines detected in an extragalactic source to date.

The relative strengths of the H2O and OH lines in NGC 4418 and Arp 220 show a striking dependence on Elower. The lowest-lying (Elower ≲ 200 K) H2O lines, especially the 212 → 101 transition at 179.5 μm (Fig. 4q), are stronger in Arp 220 than in NGC 4418, but higher-lying lines are stronger in NGC 4418 (see Fig. 13). The effect is even more clearly seen in OH (Fig. 14), where the ground-state lines (upper row in Fig. 6) are much stronger in Arp 220, the intermediate Π3/25/2 → 3/2 transitions at 84 μm (Fig. 6e) have similar strengths, and higher excited OH lines (Figs. 6f − i) are stronger in NGC 4418. This is partially a consequence of the Cextended component in Arp 220: since Weq are calculated relative to the observed (nuclear+Cextended) continuum, the high-lying lines arising from the nuclei are reduced due to the contribution by Cextended to the continuum, which is more prominent in Arp 220. In other words, the high-lying lines are more continuum-diluted in Arp 220 than in the relatively “naked” NGC 4418, which is consistent with the continuum models shown in Sect. 3. On the other hand, Cextended in Arp 220 produces strong absorption in the low-lying lines, thus increasing their Weq relative to that in NGC 4418.

Nevertheless, dilution in the continuum cannot explain the higher excitation indicated by the very high-lying lines of H2O and OH in NGC 4418. This is best seen in Figs. 13g, h, where the line fluxes are normalized relative to that of the excited 330 → 221 H2O line at 66.4 μm. The latter line (Fig. 3g) is well detected in both sources and, with Elower = 160 K, is not expected to be significantly contaminated by absorption of extended low-excitation H2O in Arp 220. Lines with Elower above 500 K are shown to produce, relative to 330 → 221 , more absorption in NGC 4418 than in Arp 220. Also, as mentioned above, the H2O 818 → 707 and 716 → 625 transitions (Figs. 3f, g) are only detected in NGC 4418 (the 808 → 717 line in Fig. 3f is contaminated by NH3, and the feature is not detected in Arp 220). Two more para-H2O transitions detected only in NGC 4418 are the 533 → 422 and 440 → 431 lines (Figs. 3a, r); the lower levels of these lines are non-backbone and thus require high columns to be significantly populated. Other H2O lines detected only in NGC 4418 are the 625 → 616 and 441 → 432 at 94.7 μm, but the Arp 220 spectrum at this wavelength shows a very broad feature, possibly due to pointing or instrumental effects (Fig. 3q).

The spectrum of NGC 4418 clearly shows two intrinsically weak H2O lines with ΔK ± 1 =  ± 3, the 431 → 404 line at 61.8 μm and the 330 → 303 line at 67.3 μm (Figs. 3e, h). All other detected H2O lines have higher transition probabilities and, with the exception of the very high-lying lines, are expected to be strongly saturated. While the ortho-H2O 67.3 μm line is also detected in Arp 220 within a wing-like feature that probably has some contribution from O 330 → 221 , the para-H2O 61.8 μm line is not detected in Arp 220. The high columns in NGC 4418 are also confirmed by comparing the absorption in the ortho/para pair 441 → 432 and 440 → 431 (Figs. 3q, r), which have similar lower level energies (≈550 K), wavelengths, and transition probabilities (A ≈ 0.15 s-1). For an ortho-to-para ratio of 3, one would expect the same ratio of 3 for Weq in the optically thin regime, but this is as low as 1.15 indicating that both lines are still well saturated.

Even higher-excitation lines of H2O, with Elower ≈ 900−1100 K, may be present in the spectrum of NGC 4418 (blue-dashed arrows in Fig. 2). Among these features, shown in Fig. 5, the most likely detection is the 726 → 615 line in Fig. 5c, all other features being marginal. In particular, the 92.8 μm feature in (f) has a non-Gaussian shape. Nevertheless, all features together appear to indicate more extreme excitation in the nucleus of NGC 4418. This is plausible, as HCN rotational lines with Elower ≳ 1075 K are detected in NGC 4418 (see below).

As discussed earlier, dilution in the continuum also lowers the Weq of the high-lying OH lines in Arp 220 relative to the values in NGC 4418; however, the most excited OH transitions, the Π1/2   9/2 → 7/2 and Π1/2   11/2 → 9/2 lines at 56 and 53 μm, respectively, are intrinsically stronger in NGC 4418 according to our decomposition of the continuum. Another effect of Cextended is also worth noting: the only OH transition observed in emission above the continuum, the Π1/2   3/2 → 1/2 doublet at 163 μm, is excited through absorption of continuum photons in the 35 and 53.3 μm doublets followed by a cascade down to the ground OH state (see Appendix II by Genzel et al. 1985), and is therefore expected to arise in the Cextended component (G-A04). The strong OH 163 μm emission in Arp 220 reflects the massive envelope around its nuclei. In NGC 4418, the doublet is weaker but still in emission, also hinting at the existence of a Cextended component around its nucleus.

2.4. O and 18OH

The spectra around relevant O and 18OH transitions in NGC 4418 and Arp 220 are displayed in Figs. 7 and 8. Close spectral features due to H3O+, NH2, C3, and CH+ are also indicated. O is detected in both sources, though in relatively low-lying lines. The highest-lying detected line is the 432 → 321 transition with Elower ≈ 270 K in Fig. 7a, as detection of the 616 → 505 transition in panel e is rather marginal. The most striking feature in Fig. 7 is the relative amount of absorption observed in NGC 4418 and Arp 220. While the absorption in the main isotopologues is stronger in NGC 4418 than in Arp 220 (Figs. 3 and 4), the opposite behavior is generally found for the absorption in the rare isotopologues. This is best seen in the 331 → 220 , 321 → 212 , 221 → 110 (showing redshifted emission in Arp 220), 313 → 202 , and 221 → 212 lines. There are, however, two spectral features nearly coincident with the 423 → 312 and 414 → 303O lines at 79.5 and 114.3 μm (panels d and h, respectively), with absorption in NGC 4418 apparently stronger than in Arp 220, but the latter is subject to an uncertain baseline.

The 18OH lines in Fig. 8, specifically the 18OH 85 and 120 μm doublets, confirm the 18O enhancement in Arp 220. The blueshifted Λ-component of the 18OH 120 μm doublet is strongly contaminated by CH+J = 3 → 2 in both sources, but the uncontaminated redshifted Λ-component, undetected in NGC 4418, is strong in Arp 220. It is also worth noting that the two 18OH 85 μm Λ-components show rather different absorption depths in Arp 220. This Λ-asymmetry is not expected to arise from radiative pumping or opacity effects, because the Λ-components of any transition have the same radiative transition probabilities, and the corresponding lower levels tend to be equally populated. Indeed, the absorption in the corresponding Λ-components of the main isotopologue at 84 μm are very similar (Fig. 6e). The 18OH 85 μm Λ-asymmetry in Arp 220 is further discussed in Sect. 3.2.3. The 18OH Λ-component at 65.54 μm in Fig. 8a is partially blended with two lines of NH2 and, together with the expected contribution by H2O+ at 65.61 μm, form a broad feature with uncertain baseline. The red Λ-component at 65.69 μm appears to be stronger in Arp 220 than in NGC 4418.

2.5. HCN

HCN is a key molecule widely studied in both Galactic and extragalactic sources. In NGC 4418 and Arp 220, the low-lying lines at millimeter wavelengths have been observed by Aalto et al. (2007a,b) and Wiedner et al. (2004). The ν2 = 11 vibration-rotation band at 14 μm was detected in both sources by Lahuis et al. (2007), and rotational emission from the upper vibrational state in NGC 4418 has been detected by Sakamoto et al. (2010). The ν2 = 11 l-type HCN lines from J = 4−6 at centimeter wavelengths were detected in absorption toward Arp 220 (Salter et al. 2008). Among the 15 extragalactic sources detected in the HCN band, the Lahuis et al. (2007) analysis indicated that NGC 4418 has the second highest HCN column. Recently, Rangwala et al. (2011) have reported the detection of several emission/absorption HCN lines with Herschel/SPIRE in Arp 220.

We report in Fig. 9 the PACS detection of HCN absorption in pure rotational transitions at far-IR wavelengths in both NGC 4418 and Arp 220. In NGC 4418, all transitions from J = 18 → 17 (Elower = 650 K) to J = 23 → 22 (Elower = 1075 K) are clearly identified. Furthermore, there are hints of absorption up to the J = 25 → 24 transition (Elower = 1275 K). In Arp 220, the J = 20 → 19 line is clearly contaminated by H3O+ and NH2, and the J = 23 → 22 one by C3 and possibly also by NH2. The J = 19 → 18 line is weak, suggesting that this transition is tracing the tail of the Spectral Line Energy Distribution (SLED). However, a clear feature is found at the wavelength of the J = 21 → 20 line, and the J = 22 → 21 line could show some marginal absorption as well. Nevertheless, we consider the identification of the J = 21 → 20 line as questionable given the weakness of the nearby rotational HCN lines and the possible contamination by other species.

thumbnail Fig. 3

H2O lines in the blue and green bands (λ < 100 μm) detected in NGC 4418 (upper spectra in each panel) and Arp 220 (lower spectra). The spectra, histograms with black solid lines, have been scaled by a factor of 0.5 in panels k) and p). Vertical dotted lines indicate the rest wavelengths of the lines relative to z = 0.00705 and 0.0181 for NGC 4418 and Arp 220, respectively. Model results are also shown. For NGC 4418, the blue and light-blue curves show the models for the Ccore and Cwarm components, respectively; for Arp 220, the blue and light-blue lines show the estimated contributions from Cwest and Ceast , and the green curve (in panel k) shows the contribution by Cextended . Red is total. Model parameters are given in Tables 1 and 2.

Open with DEXTER

thumbnail Fig. 4

H2O lines in the red band (λ > 100 μm) detected in NGC 4418 (upper spectra in each panel) and Arp 220 (lower spectra). The spectra have been scaled by a factor of 2 in panels a), k) and l), and by 3 in panel c). Model results are also shown. Color codes for the contributions of the different components are given in Fig. 3, with the yellow curve in panel q) showing the additional contribution by the halo component (Chalo) in front of the nuclear region of Arp 220.

Open with DEXTER

Fluxes derived from Gaussian fits are shown in Fig. 15 as a function of Elower, where values derived from Herschel/Spire observations by Rangwala et al. (2011) are included for Arp 220. The HCN excitation in NGC 4418 is extreme, with the SLED apparently peaking at J = 21 → 20 (Elower ≈ 900 K). The striking characteristic of the far-IR HCN lines is that they are detected in absorption against the far-IR continuum, indicating that the dust temperature and far-IR continuum opacities behind the observed HCN are high; otherwise the lines would be detected in emission. We argue in Sect. 3 that the H2O, OH, and HCN lines in NGC 4418 are tracing a high luminosity, compact nuclear source, denoted as the nuclear core (Ccore). In Arp 220, the HCN SLED as seen by Herschel/Spire (Rangwala et al. 2011) peaks at J = 16 → 15 (Elower ≈ 500 K) indicating, like H2O, more moderate excitation, but a second, higher-excitation component could be present if the 161.3 μm spectral feature is due to HCN J = 21 → 20. As shown in Sects. 3.2.4 and 3.4.1, high HCN columns are required to explain the data.

thumbnail Fig. 5

Marginally detected H2O lines in NGC 4418 (upper spectra in each panel); the corresponding lines from Arp 220 are also shown (lower spectra). Model predictions for these lines (for Ccore in NGC 4418 and for Cwest in Arp 220) are also shown (red curves).

Open with DEXTER

2.6. NH3

Ammonia is another N-bearing species widely observed in Galactic sources mostly through the pure-inversion transitions from metastable (J = K) levels at centimeter wavelengths (see Ho & Townes 1983, for a review). The rotation-inversion transitions lie at far-IR wavelengths, and many of them have been detected with ISO toward Sgr B2 (Ceccarelli et al. 2002; Polehampton et al. 2007) indicating a hot, low dense molecular layer interpreted in terms of shock conditions (Ceccarelli et al. 2002). Detections of far-IR NH3 lines in extragalactic sources were presented, together with a preliminary analysis, by G-A04 in Arp 220. In Fig. 10 we show the Herschel/PACS detection of high-excitation, far-IR NH3 lines in both NGC 4418 and Arp 220.

NH3 has two species, ortho-NH3 (K = 3n) and para-NH3 (K ≠ 3n), with an expected ortho-to-para abundance ratio of unity (Umemoto et al. 1999). All rotational (J,K) levels, except those of the K = 0 ladder, are split into 2 inversion doubling sublevels (a for asymmetric and s for symmetric), which for a given J have decreasing level energy with increasing K. Radiative transitions are only allowed within K-ladders, thus only collisions can populate the K > 1 ladders and the absorption from different K-ladders is then sensitive to the gas temperature. Within a given K-ladder, both collisions and absorption of far-IR photons can pump the J > K non-metastable levels. Since the a → s (s → a) J + 1 → J transitions of different K-ladders have similar wavelengths, the lines are crowded in wavelength and overlapping in some cases, with severe blending between the K = 0 and K = 1 ladders. The PACS domain covers the range from the (3,K)a → (2,K)s (≈170 μm, Fig. 10g) and the (3,K)s → (2,K)a (≈165.7 μm, Fig. 10f) lines, up to (10,K) → (9,K) at 50−51 μm.

Figure 10 also indicates (partial) blending with lines of other species, as H2O (panels a and b), CH+ (panel b), and H3O+ (panel f). Most (6,K)a → (5,K)s lines are strongly blended with the OH 84 μm doublet. From the comparison with the modeling described in Sect. 3, the (6,K)s → (5,K)a and especially the (7,K)a → (6,K)s lines are so blended that a pseudo-continuum is expected to be formed, with the consequent uncertainty in the baseline subtraction. As also found for H2O, OH, and HCN, NGC 4418 shows in NH3 clear indications of higher excitation than Arp 220, and the absorption in non-metastable levels relative to metastable ones is also stronger in NGC 4418 (e.g., the (4,K)s → (3,K)a group in panel d). The ortho-NH3 (8,6)a → (7,6)s is probably detected in NGC 4418, as the blended para-H2O 808 → 717 line (see also Fig. 3f) is probably less strong than the close ortho-H2O 818 → 707 line, and also because the (8,6)s → (7,6)a transition is detected. There are also some hints of absorption in the (8,5)s → (7,5)a line. In summary, NH3 lines are detected up to (8,K) → (7,K) in NGC 4418 (Elower ≈ 600 K), and up to (7,6) → (6,6) in Arp 220 (Elower ≈ 400 K).

2.7. Kinematics

In Fig. 12, we compare the HCO+J = 3−2 (green histograms) and J = 4−3 (blue histograms) line profiles toward the western (a) and eastern (b) nucleus of Arp 220 (from Sakamoto et al. 2009), with the line profiles of selected lines detected with Herschel/PACS (c-i). The HCO+ lines toward both nuclei, observed with high angular (0.3′′) and spectral (30 kms-1) resolution, exhibit P-Cygni profiles indicative of outflowing gas (Sakamoto et al. 2009). Specifically, the redshifted spectral features observed in emission have a large velocity extent of  ~500 kms-1 from the systemic velocity. This prominent redshifted HCO+ emission has its counterpart in several lines detected with Herschel/PACS: the H2O 303 → 212 line at 174.6 μm (Fig. 12c), the [O i] 63 μm transition (d), and the OH 79 μm doublet (e). The latter also shows deeper absorption in the blue Λ-component than in the red one, suggesting the occurrence of redshifted emission in the blue Λ-component as well. The line emission features are expected to be formed in gas located at the back/lateral sides (i.e. not in front of the nuclei where absorption of the continuum dominates the profile), and will therefore be redshifted, as observed, if the gas is outflowing. The velocity of the emission features (vLSR ~ 5650−6000 kms-1) is significantly higher than the redshifted velocities measured for the rotating disks around both nuclei (up to  ~5700 kms-1, Sakamoto et al. 2008), indicating that little of this emission is associated with the rotation motions. Thus we conclude that moderate excitation lines of H2O, OH, and O i trace the outflow at redshifted velocities detected in HCO+. The asymmetric shape of the OH 163 μm emission doublet, which peaks at redshifted velocities, is consistent with this scenario, as is the detection of redshifted emission features in the Herschel/SPIRE spectrum of Arp 220 in several species including H2O (Rangwala et al. 2011).

thumbnail Fig. 6

OH lines in NGC 4418 (upper spectra) and Arp 220 (lower spectra). Λ-doubling is resolved in most transitions, except for the two narrowly separated 71 μm components that are blended into a single spectral feature. Model results are also shown. For NGC 4418, the blue, light-blue, and green curves show the models for the Ccore , Cwarm , and Cextended components, respectively; for Arp 220, the blue, light-blue, green, and yellow curves show the models for Cwest , Ceast , Cextended , and Chalo , respectively. The gray curves in panel f) show the expected contribution of the H2O 625 → 514 line to the spectra around 65.2 μm. The red curves show the total from all components.

Open with DEXTER

The blueshifted velocity extent of the OH 163 and [O i] 145 μm lines, observed in emission, reaches  ~400 kms-1 from the systemic velocity (Figs. 12h–i), and could also trace outflowing gas at the most extreme velocities. The full velocity extent in these lines (≈1000 kms-1) is similar to that of CO (3–2) (Sakamoto et al. 2009). The full velocity extent of the high-lying H2O lines is significantly lower,  ≈550 kms-1. This is specifically illustrated in Figs. 12f–g for the 322 → 211 and 616 → 505  lines, which lie in the range 90−82 μm where PACS has relatively high velocity resolution (130−150 kms-1). In contrast with the OH 163 and the [O i] 145 μm lines observed in emission, which probably trace spatially extended gas, the formation of the high-lying absorption H2O lines is restricted to regions with high far-IR radiation densities that are optically thick at far-IR wavelengths, obscuring the emitting gas behind the nuclei. The absorption in these H2O lines peaks at around central velocities, indicating that the lines mainly trace gas rotating on the surface of the nuclei. However, due to uncertainties in the velocity correction due to pointing shifts (see Sect. 2), and the similarity in the blueshifted velocity extent of the HCO+ and H2O lines, we cannot rule out a significant contribution to the absorption by outflowing gas with velocities up to  ~150 kms-1. Nor can we rule out spatially extended gas inflow, similar to what is inferred and discussed below for NGC 4418, since the peak absorption in the OH 79 μm doublet is redshifted by  ≈90 kms-1.

In NGC 4418, the adopted redshift (z = 0.00705) is derived from the excited H2O and OH lines, coinciding within  ≈15 km s-1 with both the centroid of the [C ii] 158 μm line and with the redshift inferred from the stellar absorption at optical wavelengths (Sect. 2.2). No indication of outflowing gas in the nuclear region is found. The excited H2O and OH lines in NGC 4418 are indeed significantly narrower than in Arp 220 (Figs. 13 and 14), further indicating the relatively quiescent state of the NGC 4418 nucleus. On the other hand, the line shapes of the lowest-lying H2O, OH, and [O i] lines show an intringuing behavior, opposite to the outflow signatures in Arp 220: the absorption features in the ground OH and [O i] 63 μm lines are systematically redshifted by  ~100 kms-1, and the [O i] 63 μm line shows a blueshifted spectral component in emission, peaking at  ≈−55 kms-1, i.e. an inverse P-Cygni profile (Fig. 11a). The lowest-lying H2O line at 179.5 μm is also redshifted, but contamination by CH+ (2 → 1) at 179.594 μm makes the case more uncertain. Additional clues about this component come from the line shapes of the OH 84 and 163 μm doublets: the blue Λ-component of the 84 μm doublet (Elower = 121 K) shows an asymmetric shape with some redshifted excess (Figs. 6e and 11e; the red Λ-component is probably contaminated by excited NH3, see Fig. 10c), suggesting that some redshifted gas at  ~50 kms-1 is significantly excited. The kinematics implied by the OH 163 μm emission doublet are unclear due to the low signal-to-noise ratio (SNR), as one of the components is apparently blueshifted, whereas the other one peaks at central velocities (Figs. 6j, 11f, and 14). In summary, the ground-state lines of OH and [O i], together with the OH 84 and 163 μm doublets, indicate the presence of a component different from the nuclear one in both excitation and kinematics, most probably extended in comparison with the nucleus but much less massive than the Cextended component of Arp 220, and the velocity shifts suggest that this component may be inflowing onto the quiescent, nuclear region. We further investigate this scenario in Sect. 3.6.

thumbnail Fig. 7

Spectra around the wavelengths of relevant O lines in NGC 4418 (upper profiles) and Arp 220 (lower profiles). Model predictions for NGC 4418 (Ccore) with , and for Arp 220 (Cwest) with are shown in red.

Open with DEXTER

thumbnail Fig. 8

Spectra around the wavelengths of relevant 18OH lines in NGC 4418 (upper profiles) and Arp 220 (lower profiles). Model predictions for NGC 4418 (Ccore) with , and for Arp 220 with are also shown. For the 18OH 120 μm doublet, the contributions by the different components in Arp 220 (see Fig. 6) are indicated; red denotes the total of all components.

Open with DEXTER

thumbnail Fig. 9

Spectra around the frequencies of HCN lines in NGC 4418 (upper profiles in each panel) and Arp 220 (lower profiles). Model predictions for NGC 4418 (Ccore) and Arp 220 (Cwest) are included.

Open with DEXTER

thumbnail Fig. 10

Spectra around the frequencies of relevant NH3 lines in NGC 4418 (upper profiles in each panel) and Arp 220 (lower profiles). Model predictions for NGC 4418 (Ccore) and Arp 220 (Cwest) are included.

Open with DEXTER

thumbnail Fig. 11

Line profiles of the [O i] 63 and 145 μm (a and b) transitions compared with those of some OH lines (c)f)). Upper histograms, NGC 4418; lower histograms, Arp 220. Model predictions are also included. For NGC 4418, the blue curves show the combined contribution of the Ccore and Cwarm components, and the green curves show the model for Cextended . For Arp 220, see caption of Fig. 6. Red denotes the total of all components.

Open with DEXTER

thumbnail Fig. 12

Comparison between the line shapes of the HCO+J = 3 → 2 (green histograms) and J = 4 → 3 (blue histograms) transitions toward a) the western and b) eastern nucleus of Arp 220 (from Sakamoto et al. 2009), with selected lines detected with Herschel/PACS (ci). The dashed vertical line indicates our adopted systemic velocity (Sect. 2.2). The velocity scale for the OH 79 (163) μm doublet is relative to the red (blue) Λ-component, and the systemic velocity for the other Λ-component is also indicated. The velocity of the O 303 → 212 line in c) is also indicated.

Open with DEXTER

thumbnail Fig. 13

Velocity shifts (ab), line widths (cd), and equivalent widths (ef) of the H2O lines in NGC 4418 (left) and Arp 220 (right). Panels gh show the line fluxes normalized to that of the 330 → 221 transition at 66.4 μm (see Sect. 2.3 for details). Error bars are 1-σ uncertainties from Gaussian fits to the lines.

Open with DEXTER

thumbnail Fig. 14

Velocity shifts, line widths, and equivalent widths of the [O i] and OH lines in NGC 4418 (black symbols) and Arp 220 (blue symbols). Error bars are 1-σ uncertainties from Gaussian fits to the lines.

Open with DEXTER

thumbnail Fig. 15

HCN line fluxes in NGC 4418 a) and Arp 220 b). In Arp 220, the fluxes of lines with Elower up to 570 K have been taken from SPIRE-FTS data (Rangwala et al. 2011). The curves and triangles show model predictions for the Ccore component of NGC 4418 and the Cwest component of Arp 220. The dotted curve in b) shows results obtained when only collisional excitation is included in the model. The solid curve in a) shows the best fit model obtained with τ200 ≈ 1, while the dashed and dotted curves correspond to τ200 = 4 with and without gas-dust mixing, respectively.

Open with DEXTER

3. Models

3.1. Overview

As the data presented in previous sections have shown, the far-IR spectra of NGC 4418 and Arp 220 are dominated by molecular absorption, with emission in only some lines and profiles. These rich line spectra and their associated dust emission cannot be described by a single set of ISM parameters, but different lines have different excitation requirements and are thus formed in different regions of the galaxies. Our approach for both galaxies is to fit these different regions and conditions, even though they are not spatially resolved with Herschel, with the smallest possible number of parameterized components.

For NGC 4418, we find that we need (see Fig. 1a) (i) a hot component (yellow curve, hereafter Chot) that accounts for the mid-IR continuum; (ii) a nuclear core (blue curve, Ccore) that provides absorption in the high-lying lines of H2O, OH, HCN, and NH3; (iii) a warm component (light-blue, Cwarm), that provides absorption in moderately excited lines of H2O and OH as well as a significant fraction of the far-IR continuum emission; and (iv) an extended component (solid green, Cextended), which accounts for the low-lying redshifted lines of OH and O i.

For Arp 220, we need (Fig. 1b) (i) a single3 hot component (yellow curve, Chot) that accounts for the mid-IR continuum emission; (ii) the western and eastern nuclear components (Cwest and Ceast in blue and light blue, respectively), where the high-lying molecular lines are formed; (iii) the extended component (Cextended, green), which provides a significant fraction of the far-IR continuum emission and which is likely associated with moderate and low-excitation lines of H2O and OH; and (iv) an additional absorbing “halo” component (Chalo), with no associated intrinsic continuum but located in front of the nuclei, is required to fit the absorption in the ground-state lines of H2O, OH, and O i.

We use single dust temperatures for every component listed above, except for the Cextended component in both galaxies where the dust temperature profile is calculated from the balance of heating (by the inner components) and cooling (González-Alfonso & Cernicharo 1999). In our models, each single dust temperature component is attenuated by a foreground, screen-like shell, which is parameterized by its dust opacity at 25 μm, τ25,fgr. These screen-like shells are responsible for the silicate absorption features imprinted on the various modeled components in Fig. 1. The dust temperature profile in Cextended is calculated by assuming spherical symmetry and a single nuclear heating component located at the center of the modeled source.

Parameters of the continuum models, and the inferred molecular parameters, are listed in Tables 1 and 2, respectively. A sketch of the modeled sources, showing the approximate spatial scales of the different far-IR components (excluding the Chot component of both galaxies), is shown in Fig. 16. Each component, however, can be interpreted in terms of a single source, as implicitly assumed, or alternatively applied to each one of an ensemble of NC smaller clouds of radius RC that do not spatially overlap along the line of sight. The scaling between these two approaches is discussed in G-A04.

The models for Arp 220 are discussed in Sects. 3.2 (high-lying lines) and 3.3 (low-lying lines), while those for NGC 4418 are developed in Sects. 3.4 (high-lying lines), 3.5 (mid-excitation lines), and 3.6 (low-lying lines). The continuum of NGC 4418 is further discussed in 3.7.

3.2. The high-lying lines in Arp 220: the nuclear region (Cwest and Ceast)

3.2.1. H2O

Our models for Arp 220 are similar to those described in G-A04, and generated on the basis of the interplay between the continuum emission and the molecular line absorption. Since the molecular excitation in Arp 220 is high and collisional excitation alone cannot account for it, the excitation mechanism is expected to be dominated by absorption of dust-emitted photons in the nuclear region of the galaxy (G-A04, González-Alfonso et al. 2008, 2010). The molecular excitation is thus a function of the dust temperature, Td, and of NH2Ov, and in combination with the observed continuum provides clues about the general properties of the far-IR continuum source and its asociated chemistry. If the observed H2O excitation cannot be reproduced by assuming that the dust in the nuclear component has a temperature of “only” Td = 85 K (derived from the observed 25 and 1300 μm emission), foreground extinction in the far-IR is included to attenuate also the 25 μm emission from the nuclei in such a way that the far-IR is reproduced with Td > 85 K. This is Scenario 2, S2, in G-A04, based on the continuum models by Soifer et al. (1999). Evidence for such foreground extinction in the mid-IR comes from the strong silicate absorption at 9.7 and 18 μm, in the millimeter by self-absorption in CO (2 → 1) (Downes & Eckart 2007), and in the far-IR from the [O i] 63 μm line that is observed in absorption, in contrast to most extragalactic sources in which it is observed in emission.

In the present models, we have simulated the dust emission by using a mixture of silicate and amorphous carbon grains with optical constants from Draine (1985) and Preibisch et al. (1993); the mass-absorption coefficient is 550, 150, and 12.3 cm2/g of dust at 25, 50, and 200 μm, respectively, and the spectral index is β ≈ 2 in the far-IR. We have also attempted to disentangle the emission from the eastern and western nuclei in Arp 220. We have modeled the eastern nucleus (Ceast) as a sphere of diameter deast = 0.31′′ (Sakamoto et al. 2008, hereafter Sa08), radial opacity (i.e. the opacity along a radial path) at 200 μm τ200 = 2.7, and Td,east = 87 K. The emission from Ceast is attenuated by foreground dust with τ25,fgr = 1, yielding 2.5−0.02 Jy at 25−1300 μm – consistent with measurements by Soifer et al. (1999) and Downes & Eckart (2007). The corresponding SED of Ceast is shown in Fig. 1b in light-blue, and accounts for a luminosity of Least = 3 × 1011 L. For the western nucleus (Cwest) we also use as a first approach the size as derived by Sa08 in the “disk” approximation, i.e. a sphere with diameter dwest = 80 pc (0.23′′ at 72 Mpc), but allow it to vary to match the observed continuum, and leave as free parameters the dust temperature Td,west and foreground extinction τ25,fgr.

Calculations for the lines were carried out in spherical symmetry using the code described in González-Alfonso & Cernicharo (1999). The models indicate that most high-lying absorption lines are formed in the outermost shell surrounding the far-IR continuum source. This is illustrated in Fig. 17, where the absorbing flux of several lines of H2O, OH, O, NH3, and HCN, is plotted as a function of the depth of the shell where the molecules are located, which is parameterized in terms of the continuum opacity at 50 μm (τ50) measured from the surface of the far-IR source. Since the molecular abundances relative to the density of dust are uniform in these models, the column densities are proportional to τ50. Results indicate that the H2O and OH absorption is produced in a thin shell with τ50 ≈ 1, as extinction, thermalization by dust emission, as well as line opacity effects make results insensitive to the presence of molecules deeper into the far-IR source. The optically thinner O lines are mostly formed in the same shell, though the lines at long wavelengths (e.g. the 221 → 212 at 183 μm) still have significant contribution (≲25%) from the 1 < τ50 < 2 region. The case for NH3 and HCN is somewhat different, as some lines probe deeper regions. The high-lying lines of NH3 lie at λ < 100 μm and are also formed in the τ50 ≈ 1 outermost shell, but the low-lying lines at λ > 100 μm have significant contribution from 1 < τ50 < 3. The high-excitation HCN lines sample regions deeper than the other species (τ50 ≲ 4) as their excitation is sensitive to the far-IR radiation density at λ ≳ 170 μm (see Sect. 3.2.4). Based on these results, the models for the nuclei shown below use a “screen” approach, i.e. the molecules are mixed with the dust but located within a thin shell surrounding the nuclei extended up to a depth of τ50 = 2 for H2O, OH, and their 18O isotopologues, and up to τ50 = 4 for NH3 and HCN. The values of the column densities of all species below, however, are given for the τ50 = 1 outermost shell that dominates the absorption of H2O and OH, and are lower limits to the true columns through the nuclear sources. In order to estimate the abundances χX relative to H nuclei, we have normalized these column densities per unit of τ50 to a given value of NH,norm. For a mass-absorption coefficient of 150 cm2/g of dust at 50 μm and a gas-to-dust mass ratio of 100, τ50 = 1 corresponds to a column density of NH,norm = 4 × 1023 cm-2, where H refers to hydrogen nuclei in both atomic and molecular forms. This value of NH,norm has been applied to all species observed in the far-IR, and define the far-IR “photosphere” where most molecular absorption is produced.

thumbnail Fig. 16

Schematic representation of the modeled sources, showing the approximate spatial scales of the different far-IR components (see Fig. 1 and Tables 1 and 2). The Chot components in both galaxies (that account for the mid-IR spectra) are most probably associated with the nuclear regions and are not included here. The Chalo component in Arp 220 has no associated continuum, and is responsible for the absorption of the nuclear continuum (Ceast and Cwest) in the ground-state lines of H2O, OH, and O i. Although, due to its apparent low excitation, Chalo is represented here as the most extended component, its size is not constrained. The plot is an oversimplification of the actual models, where the different components are modeled separately to account approximately for non-spherical symmetry.

Open with DEXTER

Line broadening is caused by microturbulence with vtur = 60 kms-1 and a velocity shift of 130 kms-1 through the absorbing shell. The latter simulates either the presence of outflowing gas or, more generally, gas velocity gradients across the nuclear regions, allowing us to nearly match the observed linewidths. Rates for collisional excitation are taken from Faure et al. (2007); with the adopted nH2 = 5 × 105 cm-3 and gas temperature Tg = 150 K (see Sect. 3.2.4), collisional excitation has little effect on the calculated fluxes. The line models have three free parameters: the dust temperature, Td, the H2O column, NH2O, and the covering factor of the continuum source, fC.

The bulk of the H2O absorption is produced in a luminous, compact region with high H2O columns, which we identify with the double nucleus of Arp 220. A number of previous observational studies with high angular resolution have revealed that the western nucleus is brighter than the eastern one in the near-IR (Armus et al. 1995), mid-IR (by a factor of  ~3 at 25 μm; Soifer et al. 1999), and (sub)millimeter wavelengths (Downes & Eckart 2007; Sakamoto et al. 2008). Given that the excitation of H2O requires a high brightness continuum source, the observed high-lying H2O absorption is tentatively attributed to Cwest . Nevertheless, the column densities derived below are independent of whether Cwest alone is responsible for the absorption in the high-lying lines, or Ceast has a significant contribution as well, and the properties we derive for Cwest and Ceast can be more generally interpreted as shared by both nuclei.

thumbnail Fig. 17

Modeled absorbing flux in Arp 220 of several a) H2O, b) OH, c) O, d) NH3, and e) HCN lines as a function of the depth of the molecular shell measured from the surface of the continuum source, as parameterized by the dust opacity at 50 μm (τ50). Numbers in parenthesis indicate rounded wavelengths in μm. The total column of molecules is proportional to τ50, and we have adopted NH2O = 2 × 1018 × τ50 cm-2, and abundance ratios of OH/H2O = 0.5, , HCN/H2O = 0.3, and HCN/NH3 = 6. The dust temperature is 110 K.

Open with DEXTER

In the nuclear region where the H2O lines are formed, the models shown below indicate that most of these lines are strongly saturated; however, there are still some high-lying lines that are sensitive to both NH2O and Td. For these critical lines, Fig. 18 shows the observed and modeled fluxes relative to that in the 330 → 221 line at 66.4 μm (Fig. 3g). As mentioned in Sect. 2, the 330 → 221 is chosen as the normalization line because it is detected with a high SNR and, with Elower = 160 K, is not expected to be significantly contaminated by absorption of extended low-excitation H2O in Arp 220. On the other hand, the 330 → 221 transition is still low in energy as compared with the high-lying lines, so that the line flux ratios are sensitive to both NH2O and Td. Furthermore, the line flux ratios do not depend on either the size of the continuum source, or on fC. These ratios are plotted as a function of Td for different “screen-like” H2O columns, ranging from NH2O = 7.5 × 1017 to 6 × 1018 cm-2, with dashed lines indicating the measured values, and dotted lines the estimated upper and lower limits. We also include in the lowest panel the upper limit derived for the 818 → 707 line. For each Td, the foreground extinction is determined by imposing that the continuum flux density at 25 μm matches the observed 7.5 Jy from the western nucleus (Soifer et al. 1999), but this has little effect on the line flux ratios as the far-IR foreground opacities are low and, in any case, the considered lines have similar wavelengths. The minimum Td is 90 K, which is the optically thick, lower limit value derived for the western nucleus from the 860 μm submillimeter continuum by Sa08.

thumbnail Fig. 18

Modeled H2O line ratios in Arp 220 as a function of the dust temperature in the western nucleus. The dashed lines indicate the observed values, and the dotted lines the estimated upper and lower limits. In the lowest panel, the dotted line is the 3σ upper limit for the 818−707/330−221 flux ratio. Squares, triangles, circles, and crosses (only for Td,west = 90 K) show model results for “screen” H2O columns of NH2O = 7.5 × 1017, 1.5 × 1018, 3 × 1018, and 6 × 1018 cm-2, respectively.

Open with DEXTER

Results of Fig. 18 show that the line ratios scale linearly with Td. The highest sensitivity to Td is found for the 615 → 524 line, since its lower level is non-backbone (Fig. 2). Among the detected lines, the line ratios can be almost equally well reproduced with Td = 90 K and the highest NH2O = 6 × 1018 cm-2, or with Td = 140 K and NH2O = 7.5 × 1017 cm-2. However, the model Td = 140 K and NH2O = 7.5 × 1017 cm-2 yields too much absorption in the undetected 818 → 707 line, slightly favoring Td ≲ 130 K. Concerning the continuum models, if we assume that the observed high-lying H2O absorption is dominated by Cwest , and that the photosphere at 25 μm has the same size as the (sub)millimeter source, a reasonable match to the 20−35 μm SED with Td,west ≤ 110 K is not possible; for example, the model with Td,west = 90 K yields  ≈4 Jy at 25 μm with no foreground extinction and the size derived from the submillimeter (Sa08), insufficient to account for the observed 7.5 Jy at 25 μm (Soifer et al. 1999). Furthermore, any reasonable fit to the continuum requires foreground absorption at 20−35 μm, and the absolute H2O fluxes also require a continuum source larger than the submillimeter one for Td,west ≲ 120 K. However, the photosphere at 25 μm (i.e. the nuclear region with τ25 ≥ 1) may be larger and colder than the submillimeter source if heated by the central core (Sa08), because the optically thinner submillimeter emission samples warm regions that are obscured at 25 μm. For Td,west = 90 K, the required source diameter is dwest = 150 pc, yielding Lwest = 6.5 × 1011 L, while for Td,west = 130 K, dwest = 80 pc and Lwest = 8.2 × 1011 L.

In summary, a range of Td,west = 90−130 K and a corresponding “screen” column per unit of τ50 of NH2O = (60−7.5) × 1017 cm-2 are derived, with the colder (<120 K) sources yielding diameters above the (sub)millimeter observed values. The corresponding luminosities are in the relatively narrow range Lwest = (6.5−8.2) × 1011 L. The corresponding estimated H2O abundances are χH2O ~ (15−2) × 10-6 relative to H nuclei.

As the reference model shown in Figs. 35 for detailed comparison with data, we use the combination Td,west = 110 K and NH2O = 2 × 1018 cm-2, with nearly complete coverage of the source (fC = 0.8). The corresponding continuum is shown with a blue curve in Fig. 1b, with the diameter of the source (106 pc) increased by  ~30% relative to the submillimeter source, and with a foreground opacity τ25,fgr = 1.4. The data also suggests the presence of a lower excitation component to attain a better fit to the 322 → 211 , 221 → 110 , and 303 → 212 lines at 90, 108, and 174 μm, respectively. Tentatively associating this component with the eastern nucleus, we derive NH2O = 3 × 1017 cm-2 and fC = 0.4. Its contribution to the H2O SLED, shown with light-blue curves in Figs. 3 and 4, is expected to be significant only for low-lying H2O lines. Continuum and line parameters of the models are listed in Tables 1 and 2.

3.2.2. OH

The observed high-lying OH doublets in Arp 220, namely the Π3/2 9/2 → 7/2 and 11/2 → 9/2 at 65 and 53 μm, and the Π1/2 7/2 → 5/2 and 9/2 → 7/2 at 71 and 56 μm, are compared in Fig. 6 (lower spectra) with the models for the western and eastern nuclei (blue and light-blue curves). Collisional rates between OH and H2 were taken from Offer et al. (1994). As for H2O, the absorption in these lines is dominated by a source with high OH column, which we tentatively identify with Cwest . In the model of Fig. 6, we have used the same parameters as for H2O, and varied the OH column density to match the observed absorption in the high-lying lines. We derive a OH/H2O abundance ratio of 0.5−1; higher OH columns overpredict the absorption in the OH 53 μm doublet (Fig. 6h). The fit to the high-lying OH lines is satisfactory except for the underprediction of the width of the Π3/2 9/2 → 7/2 65 μm doublet, which is probably contaminated by relatively weak lines of H2O (gray curve in Fig. 6f), NH2, and H2O+. Using the same source sizes as for H2O, we find a similar OH/H2O ~ 0.5−1 ratio for Td = 90−130 K. An additional extended region is required to fit the low-lying OH lines, which is discussed in Sect. 3.3.

3.2.3. HO and 18OH

Although the detected O lines arise from low-lying levels, we find that these absorption features are mainly produced toward the nuclear region with the highest column densities. Figure 7 shows the best fit for the O lines using the same reference model as for the main isotopologue, i.e. Td = 110 K. The best fit O column density is  ≈3 × 1016 cm-2, implying an O-to-O ratio as low as  ~70, and the upper limit for this ratio is estimated to be  ≈95. The O-to-O ratio slightly depends on the dust temperature, and a value of  ≈100−150 is found for the model with Td = 90 K, though the overall fit is in this case less satisfactory. Thus our data suggest an enhancement of 18O in Arp 220 relative to the solar value of  ~5. This is confirmed by the models for 18OH shown in Fig. 8, where the same 16OH-to-18OH ratio of 70 is used with Td = 110 K. The 16OH-to-18OH ratio would also increase up to 120−150 K for Td = 90 K. The 18OH 65 and 85 μm lines are nearly reproduced, but not the asymmetry in the 18OH 85 μm doublet (Fig. 8b). Similar intensity asymmetries of  ≈40% have been also observed in most OH Λ-doublets by Goicoechea et al. (2011) toward the Orion bar, where the lines are observed in emission. As Goicoechea et al. (2011) indicate, this is probably due to asymmetries in the collisional rates between OH and para-H2 (see also Offer et al. 1994). In our absorption lines, the asymmetry would arise from the collisional excitation of 18OH in the ground Π3/2   J = 5/2 → 3/2 transition, which favors the J = 5/2+ sublevel where the 85.1 μm component arises. Since collisions with ortho-H2 quench the asymmetry, the gas is not very warm but should be dense enough so that collisional excitation with para-H2 competes efficiently with the excitation induced by the radiation field. The asymmetry is not observed in the main isotopologue (Fig. 6e), indicating that either the higher column of OH efficiently mixes the populations of the sublevels, or that OH is tracing warmer gas with increased contribution by collisions with ortho-H2. However, our model with nH2 = 5 × 105 cm-3 and Tg = 150 K does not reproduce it, and several tests with higher density and lower Tg were also unsuccesful. Further, since the ground-state 18OH 120 μm doublet is not reproduced with the warm component alone, additional absorption by the Cextended and Chalo components described below, with the same 16OH-to-18OH ratio of 70, is included (Fig. 8c).

3.2.4. HCN

Our models for the western nucleus of Arp 220 can also reproduce the fluxes of the HCN lines reported by Rangwala et al. (2011) and the J = 18 → 17 and J = 19 → 18 lines detected with PACS (Figs. 9 and 15). For HCN, both collisional excitation and absorption of far-IR photons in the highest-lying lines are required to reproduce the observations. At long wavelengths (i.e. relatively low − J lines), collisional excitation in a warm and dense region is the primary excitation mechanism. For moderate dust opacities, these lines would be detected in emission above the continuum. If collisions are able to excite the molecules up to the rotational level where the dust emission becomes optically thick and the far-IR radiation becomes strong, absorption of far-IR photons in high-J lines continues to excite the molecule to higher-lying levels and the corresponding HCN lines are observed in absorption. Thus the transition from emission to absorption lines is a measure of the dust opacity and temperature of this component, and the high-lying HCN lines detected in absorption are tracing high far-IR radiation densities as well as a warm-dense environment with high HCN column densities.

A grid of models were generated by varying nH2 in the range 105−3 × 106 cm-3 and Tg between 110 and 250 K. For each value of nH2 − Tg, both NHCN and fC were varied to fit the observed SLED. In all these models, Td = 110 K was adopted. As discussed in Sect. 3.2.1, the high excitation HCN lines can be formed deeper into the dusty, continuum source than lines from other species (Fig. 17), and we extend our HCN shell up to a depth of τ50 = 4. Rates for HCN collisions with He were taken from Dumouchel et al. (2010), and scaled for the reduced mass of the HCN-H2 system to obtain the expected collisional rates with H2. Our most plausible results were found for nH2 = (4−8) × 105 cm-3 and Tg ~ 150 K, for which NHCN = (6−4) × 1017 cm-2, respectively. NHCN is twice these values for Tg = 110 K. Decreasing nH2 below 2 × 105 cm-3 would involve HCN columns above 1018 cm-2, implying extreme HCN abundances above 2.5 × 10-6 that we consider unlikely. The increase of nH2 above 106 cm-3 involves lower HCN columns, but the relative absorption in the J = 12 → 11 line becomes somewhat underpredicted and the value of fC (or the size of the continuum source) is above that derived for H2O. Results for Tg = 150 K and nH2 = 4 × 105 cm-3, with the resulting NHCN = 5.6 × 1017 cm-2, are shown in Figs. 9 and 15. The involved far-IR continuum source has the same size as for H2O and OH. Similar values of NHCN = (3−6) × 1017 cm-2 are obtained for Td = 130−90 K, though the sizes of the continuum and HCN sources are in these cases higher and lower than for Td = 110 K, respectively, and similar to those found for H2O and OH.

In our most plausible models, the HCN/H2O ratio is 0.1–0.4 (Table 2), and χHCN = (1−2) × 10-6. In Fig. 15, the calculated HCN SLED is compared with results for a model that neglects the radiative excitation by dust (dotted curve), showing that the dust has a flattening effect on the predicted SLED. Less than 10% of the total column is stored in levels associated with the observed lines; the highest populated levels are J = 5,6. Nevertheless, owing to the high continuum opacity and dust temperature in the nuclear region, as well as to self-absorption in the outermost shells of the HCN region, all HCN lines above J = 6 → 5 are predicted in absorption. The emission in the J = 6 → 5 line (Rangwala et al. 2011) is not reproduced, probably indicating that the line is formed in a colder and more extended region associated with relatively weak continuum. On the other hand, the model in Fig. 15 accounts for the observed absorption up to J = 19 → 18 but fails to explain the absorption coincident with the J = 21 → 20 transition at 161.3 μm. If this feature is due to HCN, a region as warm as that proposed for NGC 4418 (Sect. 3.4.1) would be involved.

Our models also include the HCN excitation through the pumping of the vibrationally excited ν2 = 11 state, but we find the effect negligible for the ground-vibrational state lines. Evidently, this is a consequence of the low Td used in our modeling. Rangwala et al. (2011) have proposed that the pumping of the ν2 = 11 state dominates the HCN excitation, requiring a dust temperature of Td > 350 K. If not completely extincted, this hot component would show up at mid-IR wavelengths, and there is indeed a hot component (Chot) in our continuum modeling with Td = 400 K (yellow curve in Fig. 1b) attenuated by AV ≈ 60 mag, which yields an unattenuated flux at 14 μm of  ≈2 Jy -twice as observed. However, Chot has an effective diameter of only  ≈5 pc, while the observed HCN absorption must be produced over spatial scales of  ~100 pc because, otherwise, the far-IR HCN absorption would be negligible. Thus, if the HCN molecules are exposed to the radiation of Chot , this radiation will be diluted and it remains unclear if it can significantly affect the populations. As Rangwala et al. (2011) suggest, the alternative is that there is a hot component much more prominent, but very extincted, in such a way that it does not show up in the observed SED. The occurrence of this component is plausible, but the HCN molecules responsible for the absorption are in front of, and thus apparently exposed to the non-diluted, unextincted radiation field characterized by a “moderate” Td in the range 90–130 K for Cwest , as the continuum models indicate. The associated dust is very likely optically thick at 14 μm and thus would obscure the emission from the above buried hot component. Our models for the nuclear region are, however, structureless, and more refined models are required to clarify the role of the ν2 = 11 pumping for these lines. Finally, we note that our derived HCN column is a factor of 20 higher than that derived by Lahuis et al. (2007) from the 14 μm absorption band; as the authors state, however, their analysis is based on a covering factor of unity of the mid-IR continuum and their columns are then lower limits.

3.2.5. NH3

Our model for NH3 is shown with red lines in Fig. 10 and uses the same parameters as for HCN: nH2 = 4 × 105 cm-3, Tg = 150 K, Td = 110 K, and the same source size. The model treats line overlaps, and uses the collisional rates from Danby et al. (1988). The derived column density per unit of τ50 is NNH3 ≈ 1.5 × 1017 cm-2, i.e. HCN/NH3 ~ 4 and χNH3 ≈ 4 × 10-7. The best fit for all lines is found for Tg = 150 K, as for Tg = 250 K the lower-lying (3,K) → (2,K) lines would be underestimated. There is still some missing flux in these lines at Tg = 150 K, indicating possible low-level absorption in a more extended region. Within a given K-ladder, excitation through absorption of far-IR photons dominates over collisional excitation. The data are consistent with an ortho-to-para ratio of 1, indicating a formation temperature of at least  ~30 K (Takano et al. 2000). Similar columns are obtained for Td = 90 and 130 K, with the appropiate source sizes as derived for H2O, OH, and HCN. Half of the derived NH3 column is stored in the metastable (J = K) levels, giving τ50 = 1 metastable columns in agreement with the value inferred by Ott et al. (2011) from the inversion lines for Tex = 50 K (NNH3 = 0.84 × 1017 cm-2).

3.3. The low-lying lines in Arp 220: the extended and halo components (Cextended and Chalo)

3.3.1. H2O

In Arp 220, the absorption in the low-lying 321 → 212 (Fig. 3k), 221 → 110 (Fig. 4b), 303 → 212 (Fig. 4p), and especially the ground-state 212 → 101 (Fig. 4q) lines, require a more extended, low-excitation region (Cextended, G-A04). This is consistent with the continuum models, as the nuclei cannot account for the bulk of the far-IR emission between 50 and 300 μm, and an extended component heated in part by radiation from the nuclei, but with a possible contribution from extended star formation, is invoked (green curve in Fig. 1b). In our models, Td in Cextended is calculated from the balance between heating (by the far-IR emission from the nuclei) and radiative cooling, attaining values between 40 and 90 K. In the model, Cextended extends up to an uncertain diameter of dextended = 650 pc, has a column of NH ~ 8 × 1023 cm-2, an average density of nH ~ 103 cm-3, and a mass of M ~ 3 × 109 M (Table 1). Most of its luminosity (≈7 × 1011 L) is likely reemission, rather than intrinsically generated.

Table 1

Parameters of the continuum models.

Table 2

Derived H2O column densities and abundances, column density ratios, and covering factors of the continua.

With NH2O = 3 × 1016 cm-2 and χH2O ≈ 4 × 10-8, the H2O line absorption due to Cextended is shown with green curves in Figs. 3 and 4 only for the lines mentioned above. We have used for this component a “mixed” approach (i.e. dust and molecules coexist) because of the moderate far-IR opacities. We expect reemission in the 303 → 212 and 322 → 313 lines (G-A04); in the former line reemission is indeed observed but redshifted (Sect. 2). On the other hand, a transition zone between the nuclear region and Cextended is traced by the H2O lines in emission at λ > 200 μm detected by Rangwala et al. (2011), as will be discussed in a future paper.

The depth of the ground-state 212 → 101 H2O line at 179.5 μm, however, is not quite reproduced even with the absorption by Cextended . The line is probably contaminated by CH+ (2 → 1), but its contribution is not expected to completely fill the observed absorption. This is not surprising, as foreground low-excitation molecular absorption of the far-IR emission from the nuclei is not included in the calculations of Sect. 3.2. Thus we have also included a low-excitation component, Chalo , with no intrinsic continuum but located in front of the nuclei, so that it absorbs the far-IR emission from the nuclei. This was denoted as the “halo” in G-A04, and it is probably a combination of a part of Cextended located in the sight line toward the nuclei, together with more widespread, foreground, low-excitation gas. Although, it is likely that Chalo is relatively extended due to its low apparent excitation, we note that its extent is not constrained and it could also include gas close to the nuclei. The H2O column of this component is uncertain due to line saturation, but it is consistent with NH2O ~ (5−10) × 1015 cm-2 (Table 2). The absorption in the [O i] 63 μm line (Fig. 11a) is also expected to be produced in low-excitation gas, which we identify with the Chalo component. From the observed Weq ≈ 45 kms-1 (Fig. 14) corrected for the estimated fraction of the nuclear 63 μm continuum (36%), we obtain NO0 ~ 2.5 × 1019 cm-2. For an oxygen abundance of χO0 = 2.5 × 10-4 (Cartledge et al. 2001), Chalo is characterized by NH ~ 1023 cm-2 and τ25,fgr ~ 1. The corresponding H2O abundance is χH2O ~ 0.5 × 10-7 relative to H nuclei, similar to the value found in Galactic translucent clouds (Plume et al. 2004). The inferred NH ~ 1023 cm-2 and 25 μm continuum opacity of  ~1 are close to the values obtained from the fit to the continuum for the absorbing layer in front of the nuclei and of the hot component (Table 1). While the column density across the Chalo component is only accurate within a factor 2, the above comparison suggests that Chalo is the origin of the foreground extinction of the mid-IR nuclear emission.

Figure 19b shows the predicted-to-observed flux ratios of the detected lines for the full reference model including Cwest , Ceast , and Cextended . Values for the contaminated 717 → 606 and 212 → 101 lines are not plotted. Only two lines, the 431 → 422 (147 μm) and 615 → 606 (104 μm) show departures larger than 50%, but the small predicted-to-observed value of the latter probably indicates that the feature is contaminated by OH+ (Sect. 2). Among the undetected lines, the absorption in the 533 → 422 and 413 → 404 lines at 53.1 and 187.1 μm (Figs. 3a and 4o) are too strong in all relevant models.

thumbnail Fig. 19

Predicted-to-observed H2O flux ratios in NGC 4418 (upper panel) and Arp 220 (lower panel) for the models including all components. See Sect. 3 for details.

Open with DEXTER

3.3.2. OH

The far-IR spectrum of H2O has only one ground-state line at 179.5 μm. Since this line is strongly saturated, the species is better suited to describe the region of highest molecular excitation. OH has, however, within the PACS wavelength coverage three ground-state doublets with different strengths at 119, 79, and 53.3 μm (Fig. 2), allowing us to better characterize the Cextended and Chalo components. This is illustrated in Figs. 6a–c, where the highly excited OH in the nuclei is expected to produce weak absorption in these ground-state lines, thus requiring a more extended region where relatively low-excited OH absorbs a significant fraction of the far-IR continuum of the galaxy. Two more OH transitions also trace Cextended : the Π3/2 7/2 → 5/2 doublet at 84 μm (Fig. 6e), and the Π1/2 3/2 → 1/2 doublet at 163 μm. The former line, with Elower = 120 K, is also too strong to be reproduced with the model for the nuclei, and should be significantly excited in Cextended . The 163 μm doublet, the only one observed in emission, is excited in Cextended through absorption of continuum photons in the ground-state 53.3 and 35 μm lines, followed by cascade down to the ground OH state (Genzel et al. 1985), and thus traces large spatial-scale gas illuminated by the strong far-IR continuum (G-A04). We indeed expect the 163 μm doublet to be in absorption toward the nuclei (Fig. 6j), and thus Cextended is required to produce emission in the line. The above pumping mechanism also has observational effects on the ground-state 79 μm line: since every 163 μm photon should be accompanied, through downward cascade, by a 79 μm line photon, the emission in the 163 μm doublet ensures emission in the 79 μm transition as well. This is indeed observed in Arp 220 (Figs. 6b and 11d), where the 79 μm doublet shows an emission feature at the most redshifted velocities of 300 kms-1 with respect to the systemic one; at lower velocities, one would expect a combination of emission and absorption with the latter dominating for suitable geometries and/or sufficiently large amounts of OH in front of the far-IR continuum source. With 163 μm absorption in front of the nuclei, and emission from more extended regions, the 163 μm doublet is expected to be redshifted in outflowing gas, as observed in Arp 220 (Fig. 6j).

Table 3

H2O line equivalent widths, fluxes, and velocity shifts.

Table 4

[O i] and OH line equivalent widths, fluxes, and velocity shifts.

Table 5

HCN line equivalent widths, fluxes, and velocity shifts.

Table 6

NH3 line equivalent widths and fluxes for relatively isolated spectral features.

Table 7

O line equivalent widths and fluxes.

Table 8

18OH line equivalent widths and fluxes.

In our specific model for Arp 220, the required OH column and abundance in Cextended are NOH ≈ 1017 cm-2 and χOH ≈ 1.3 × 10-7, much lower than towards the nuclear region and favoring an OH/H2O column ratio of  ~3. This ratio reverses the value found in the nuclear region where OH/H2O is  ~0.5. The Cextended component then accounts for the 163 μm line, though the redshift is not reproduced as no outflow is included in this component, and for the absorption at 84 μm. Similar to H2O, however, the full absorption in the ground-state 119, 79, and 53.3 μm doublets is not reproduced; in particular, the absorption in the 79 μm line is predicted to be very weak. We again find that the missing absorption can be attributed to Chalo , requiring a column of NOH ~ 8 × 1016 cm-2. The contribution of this component to the absorption in the ground-state 119, 79, and 53.3 μm doublets is shown with yellow curves in Figs. 6a–c. The OH/H2O column density ratio in Chalo is estimated to be  ~10, but the relative OH and H2O columns in the Cextended and Chalo components are uncertain due to the difficulty of disentangling the two components.

The ground-state 35 μm doublet of OH (Π1/2 → Π3/2   5/2 → 3/2) has been observed with ISO by Skinner et al. (1997); correcting their continuum flux density at 35 μm (62 Jy) to the lower value of 40 Jy indicated by the Spitzer IRS spectrum in Fig. 1b, the observed absorbing flux in the 35 μm doublet is  ≈1.3 × 10-12 erg s-1 cm-2. The joint contribution of Cwest , Ceast , Cextended , and Chalo predicts a comparable absorbing flux of  ≈1.1 × 10-12 erg s-1 cm-2, dominated by Chalo (G-A04).

3.4. The high-lying lines in NGC 4418: the nuclear core (Ccore)

3.4.1. HCN and H2O

Similar to Arp 220, NGC 4418 shows strong mid-IR silicate absorption (Fig. 1, see also Spoon et al. 2001) and absorption in the [O i] 63 μm line, indicating important mid-IR extinction. We have thus followed a similar analysis as for Arp 220, based on the combination of the continuum and line absorption. In NGC 4418, line broadening is purely modeled as microturbulence with vtur = 60 kms-1.

The extreme excitation of H2O and HCN provides the most stringent requirements for the dust temperature (Td), gas temperature (Tg), density (nH2), and column densities in the nuclear region. The high-lying lines of these species arise in a compact component that we denote as the “nuclear core” (Ccore). Preliminary models for HCN in the Ccore component of NGC 4418 are shown in Figs. 9 and 15, with the HCN shell extended up to a depth of τ50 = 4. We generated several models that included/excluded excitation through collisions, absorption of far-IR radiation, and pumping through the ν2 = 11 state, and found that HCN is primarily excited through collisions, with the far-IR emission significantly modifying the distribution of populations in the highest lines. With the SLED apparently peaking at J = 21 → 20, the Ccore model requires nH2 ~ 3 × 106 cm-3 and Tg in the most probable range of 220−250 K, significantly higher than in Arp 220. Decreasing nH2 by a factor of 2 would increase Tg to  ≈300 K. The high density and Tg values are also required to reduce the otherwise too strong OH absorption in the ground-state lines, to account for the NH3 observations described below, and to reproduce the far-IR CO emission (discussed in a forthcoming paper). In spite of the high collisional excitation, the HCN lines are detected in absorption, indicating that the dust in Ccore is warm and at least a fraction of Ccore is optically thick at far-IR wavelengths. A reasonable fit to the HCN lines is obtained with Td = 140−150 K and τ200 ≳ 1 for the continuum. With τ200 ≈ 1 (solid curve and triangles in Fig. 15), we expect the rotational lines Jupper ≤ 14 to be in emission. The strength of the emission lines would decrease if τ200 ≈ 4 (dashed curve), and further weaken with a screen model for HCN (dotted curve). The inferred HCN column per unit of τ50 is NHCN = (6−7.5) × 1017 cm-2, i.e. χHCN = (1.5−2) × 10-6. As for Arp 220, we find that the pumping of the ν2 = 11 state has little effect on the excitation and strength of the detected high-J ground-vibrational state lines, as expected from our moderate value for Td (Sakamoto et al. 2010).

The minimum diameter of Ccore , d = 16.5 pc, is obtained by assuming fC = 1. The corresponding luminosity is  ≈6 × 1010 L, about 60% of the entire luminosity of the galaxy. The Ccore component cannot be much bigger or warmer as its luminosity would exceed the bolometric luminosity of the galaxy (assuming spherical symmetry for the emitting radiation). We also attempted to fit the HCN SLED with lower Td in the range 110−140 K. At 130 K, the source size is increased by 18% relative to the 150 K case, the Ccore luminosity is decreased by 20%, and the J = 25 → 24 line is underestimated; thus our modeling favors Td ≳ 130 K. With Td = 150 K, most energy is emitted at wavelengths shorter than 40 μm, but the flux density at 150−190 μm is still 4.5−3 Jy, and this is the continuum absorbed by the HCN. In Fig. 1a, the SED of Ccore (blue curve), with an adopted d = 20 pc, is attenuated by foreground dust with τ25,fgr = 1.5. The corresponding covering factor for HCN is fC ≈ 0.7. With τ200 = 1−2, the radial column through Ccore is NH ~ (0.5−1) × 1025 cm-2, sufficient to strongly attenuate the X-ray emission from an AGN located at the center.

The Ccore model was applied to H2O, and results are shown in Figs. 3 and 4 (blue curves). The highest excited H2O lines are the least saturated ones, which makes them most sensitive to the column density, and we derive NH2O in the range (3−6) × 1018 cm-2. The H2O abundance relative to H is as high as χH2O = (0.5−1.5) × 10-5, and the corresponding HCN/H2O ratio in Ccore is 0.1−0.3. The H2O models also confirm the very warm (Td = 140−150 K) source of far-IR emission, though the dust mixed in the H2O shell has a somewhat lower Td = 130 K to avoid strong emission in the 624 → 615 line at 167 μm (Fig. 4n). H2O excitation through collisions has two main effects: first, it increases the excitation of the low-lying lines among the lowest  ≈8 levels of ortho and para-H2O (over that of pure radiative pumping), providing a better fit to the observations; second, it also significantly affects some high-lying lines that, although radiatively pumped, are populated from the reservoir of H2O in the low-lying levels. This combination of increased collisional excitation in the low-lying levels, coupled to the radiative pumping to upper levels, affects mainly the 818 → 707 and 615 → 524 transitions, which are strengthened by  ~40% relative to models where collisional excitation is neglected.

Our best fit model for H2O uses τ200 = 1−1.5; a higher value (e.g. τ200 ≈ 4), still compatible with HCN, would produce too strong absorption in the H2O lines at long wavelengths (≳150 μm, specifically in the 303 → 212 transition in Fig. 4p). On the other hand, the covering factor for H2O is slightly higher than for HCN. These points could indicate that the very excited HCN is more concentrated than H2O (and mostly OH) toward the warmest region of Ccore . Figures 319 indicate that the model for Ccore is satisfactory for a number of H2O lines, and Fig. 5 shows that the model with NH2O = 5 × 1018 cm-2 predicts absorption comparable to that observed in most of the marginally detected H2O lines.

3.4.2. OH

The high-lying OH doublets in the Π3/2 ladder at 65 and 53 μm, and in the Π1/2 ladder at 71 and 56 μm, are compared in Fig. 6 (upper spectra) with the model for Ccore (blue curves). In this model we have also adopted the same parameters as for H2O, and the fit to OH shown in Fig. 6 is obtained with an OH abundance decreased by factor of 2.5 relative to that of H2O. The OH 65 and 71 μm doublets are somewhat underestimated. The OH/H2O ratio in Ccore of NGC 4418,  ~0.4, is similar to that found in the nucleus of Arp 220. Like H2O, the model for OH also favors τ200 ≈ 1 to avoid too strong absorption in the 163 μm doublet, and the covering factor for OH is higher than for HCN and H2O (Table 2). Similar to Arp 220, all other OH lines have important contribution from a more extended region, and are discussed below in Sects. 3.5 and 3.6.

3.4.3. NH3

The Ccore model for NH3, shown with red lines in Fig. 10, uses the same parameters as derived above for HCN: nH2 = 3 × 106 cm-3, Td = 150 K, and Tg = 200 and 250 K, but results for these two Tg values were found to be very similar. The NH3 data are mostly compatible with τ200 ≈ 1. The derived column is NNH3 ≈ 2.5 × 1017 cm-2, i.e. HCN/NH3 ≈ 3 and χNH3 ≈ 6 × 10-7. In comparison with the model for Arp 220, the higher Td in NGC 4418 translates into a higher excitation of non-metastable levels along a given K-ladder, and the higher Tg implies higher excitation across K-ladders. The former effect is responsible for the asymmetry between the 124.6 and 124.9 μm features in Fig. 10d, which is not fully accounted for by the model; nevertheless, the effect is sensitive to the overlap between the K = 0 and K = 1 lines. On the other hand, the model accounts for the absorption in the (8,6)s → (7,6)a line and, with the contribution of H2O, for the feature at 63.47 μm, though the marginal absorption in the (8,5)s → (7,5)a transition is not quite reproduced. As for Arp 220, the (3,K)a → (2,K)s feature (panel g) is not fully reproduced, suggesting a more extended, lower-excitation component for the low-lying lines.

3.4.4. H218O and 18OH

Relative to the main isotopologues, the O and 18OH lines in NGC 4418 are weaker than in Arp 220. We have modeled these lines in NGC 4418 with the same nuclear Ccore model as for H2O and OH, and modeled results are compared with observations in Figs. 7 and 8. For O, the model for the Ccore component (red curves in Fig. 7) that reproduces the 331 → 220 and 321 → 212 lines uses a H2O-to-O ratio of 500. While the model predictions are consistent with most observed O features, the relatively strong O 423 → 312 line at 79.5 μm (Fig. 7d) is underestimated by the model. On the other hand, the 18OH lines are only marginally detected in the spectrum of NGC 4418. Our model (red lines in Fig. 8) also uses an OH-to-18OH ratio of 500, with the result that the undetected red 120.2 μm Λ-component and the 18OH 85 μm doublet are overpredicted. Still, the low SNR 18OH lines at 65 and 85 μm are affected by an uncertain baseline; owing to additional uncertainties in the OH column and distribution, we estimate a lower limit for the OH-to-18OH ratio of 250. In summary, the 16O-to-18O ratio in NGC 4418 is estimated to be  ≳250, with a best fit overall value of  ~500.

3.5. The mid-excitation H2O lines in NGC 4418: the warm component (Cwarm)

Figures 3 and 4 show that the model for Ccore underestimates the absorption in some mid- and low-lying lines of H2O (330 → 221, 321 → 212 , 322 → 211 , 221 → 110 , 414 → 303 , 313 → 202 , and 212 → 101), and Fig. 1 indicates that the SED of the attenuated Ccore cannot account for the bulk of the far-IR emission. Both points indicate the occurrence of another nuclear component that we denote as the warm component, Cwarm .

The parameters for Cwarm are more uncertain than those for Ccore . Assuming that the continuum emission from Cwarm is attenuated by the surrounding extended component (Cextended) described below, the dust temperature is  ~85−110 K and d ~ 30−40 pc (Table 1). The model for the continuum and H2O lines of Cwarm is shown with light-blue curves in Figs. 14. The H2O lines mentioned above can be reproduced with NH2O ~ 5 × 1016 cm-2, much lower than toward Ccore . With an estimated τ200 ≈ 0.5, the model for Cwarm produces emission in some lines at λ ≳ 130 μm. The H2O abundance is several    ×10-7.

The model for Cwarm is applied to OH in Fig. 6 (light-blue curves), with OH/H2O = 0.4. The main effect of Cwarm is to increase the absorption in the 119 and 84 μm doublets, and to generate reemission in the 163 μm doublet. Nevertheless the joint contribution of the Ccore and Cwarm components predicts the 163 μm doublet in absorption, in contradiction to Fig. 6j, thus indicating the presence of a more extended component described below.

3.6. The low-lying lines in NGC 4418: the extended component (Cextended)

Figure 11 compares the [O i] 63 and 145 μm lines and the 119, 79, 84 and 163 μm OH lines in NGC 4418 and Arp 220, emphasizing the different kinematics found in both sources. In Arp 220, the OH 163 μm doublet, and the emission features in the [O i] 63 and OH 79 μm transitions, are redshifted relative to the systemic velocity, and thus reveal outflowing gas though affected by foreground absorption. The broad [O i] 145 μm line supports this kinematic scenario. In NGC 4418, however, the line profiles are narrower than, and in some aspects opposite to those in Arp 220. The main spectral characteristics of these lines in NGC 4418, as discussed in Sect. 2.7, are: (i) the [O i] 63 μm absorption is redshifted by  ≈110 kms-1, and an emission feature is detected at blueshifted velocities; (ii) the absorption in the OH 119 and 79 μm lines is also redshifted by  ≈100 kms-1, with some indications of higher redshift in the 119 μm than in the 79 μm doublet (Fig. 14); (iii) a redshifted asymmetry is seen in the red Λ-component of the 84 μm doublet; (iv) the 163 μm doublet, with relatively low SNR, is detected in emission with one of the components apparently blueshifted with a velocity comparable to that of the [O i] 63 μm emission feature, though the other one peaks at central velocities. The lowest-lying H2O line at 179.5 μm is also redshifted, but it is contaminated by CH+ (2 → 1).

On the other hand, the high-lying H2O and OH lines in NGC 4418 do not show indications of any outflow, but peak at the rest-frame velocity as derived from for the optical absorption/emission lines. Only the ground-state [O i] and OH lines are redshifted, together with some indications of a lower redshift in the excited OH 84 μm line. This indicates that this kinematic component is tracing a relatively low-excitation component around the galaxy nucleus. Near-IR observations (Evans et al. 2003) do not show evidence for a nearby “contaminating” companion where these lines could be formed (the closest galaxy, VV 655, is at  ~25 kpc), but a linear, compact structure with fainter structures extending radially out to  ~1′′ (140 pc) in radius is apparent. This morphology may suggest that the redshifted spectral features are formed in (the inner region of) these extended streamers, thereby suggesting the possibility of an envelope inflowing onto the nucleus of NGC 4418.

We first constrain the column of the redshifted absorbing component in front of the nucleus by calculating the column of O i from the Weq of the 63 μm absorption feature. Assuming that the line is optically thin at all velocities, that the shell covers the whole far-IR continuum source at 63 μm (43 Jy), and ignoring reemission in the line, we derive NOi = 5.3 × 1018 cm-2. This is a lower limit, as models described below indicate that there is significant reemission from the envelope (i.e. from Cextended) in both the 63 μm continuum and in the line; we estimate from the models a more accurate NOi = (1.4−1.8) × 1019 cm-2. Assuming a gas-phase O i abundance typical of translucent clouds (χOi = 2.5 × 10-4 relative to H, Cartledge et al. 2001; Lis et al. 2001), NH ≈ 6.5 × 1022 cm-2, or Av ~ 30 mag. Since oxygen depletion may be enhanced for those columns (Bergin et al. 2000), the allowed range is Av ≈ 30−60 mag. This extinction is at least half of that required to account for the silicate absorption at 10 μm.

One important parameter for the inflowing scenario is the amount of gas that is inflowing on the lateral and back sides of the nucleus. In order to constrain the geometry, calculations in spherical symmetry are performed and compared with data. The model for this extended component (Cextended in Tables 1 and 2) simulates a spherical envelope with inner and outer radii of Rinner ≈ 20 and Router ~ 100 pc (0.15′′ and 0.7′′) inflowing onto the warm nucleus. The value of Rinner is chosen to match the estimated outer radius of Cwarm . The line profiles are best matched with a density profile that varies as  ~r-2, with a highest value of nH = 2.5 × 103 cm-3 in the neighborhood of the nucleus. At Router, nH ≈ 102 cm-3, and thus the value for the outer radius is not critical. While the [O i]63 μm absorption can be reproduced by either a turbulent shell inflowing with uniform v = −110 kms-1, or by assuming a radial velocity gradient across the envelope, the OH shifts in Fig. 14 suggest a progressive gas deceleration with decreasing distance to the galactic nucleus, with the thickest 119 μm line, tracing the outermost regions, showing the highest Vshift, and optically thinner (79 μm) or excited (84 μm) lines peaking closer to the rest velocity. We thus have empirically adopted a velocity field with −30 kms-1 at Rinner and dv/dr = −2.1 km s-1 pc-1 (i.e. −200 kms-1 at Router), roughly simulating gas in nearly free-fall at the outermost radius, and subsequent deceleration due to radiation pressure (see Sect. 4.2.3) and/or interaction with orbiting gas, with gas approaching the systemic velocity in the neighborhood of the nucleus.

Our model for OH, shown in Fig. 11, includes both the nuclear component (the combined emission of the Ccore and Cwarm components is shown with blue lines) and the inflowing component (Cextended, green lines). An interesting feature of the line profiles is that, at central velocities where the excited lines show deep absorption (and the OH column densities are high), the absorption in the ground 119 and 79 μm lines is relatively weak. The model for the nucleus (Ccore+Cwarm) predicts too much absorption in these lines even adopting the high density of 3 × 106 cm-3 and Tg = 200 K for Ccore . Nevertheless, a reasonable match to the profiles involves partial cancellation between the absorption toward the continuum source and emission (through direct scattering at 119 μm, and pumping through 53 and 35 μm for the 79 μm transition) from Cextended around the nucleus. This process also accounts for the emission in the 163 μm doublet, as every 163 μm photon should be accompanied, through downward cascade, by a 79 μm line photon. In the model, the 163 μm components are blueshifted by −45 kms-1, but the observed shift is unclear as discussed in Sect. 2.7. On the other hand, the adopted velocity gradient predicts absorption at  ~−50 kms-1 in the 84 μm excited line, nearly matching the observed asymmetric 84 μm line shape. The model for OH is then consistent with an extended envelope and favors the scenario of an inflow extended on the lateral sides, though the lack of blueshifted emission in one of the Λ−components of the 163 μm doublet leaves the case inconclusive. The OH column across Cextended is NOH ≈ 1.5 × 1016 cm-2, and χOH ~ (1.5−3) × 10-7. From the absorption in the H2O 212 → 101 line in Fig. 4q, which is contaminated by CH+, we estimate an uncertain NH2O ~ 2 × 1015 cm-2.

The model for OH has been applied to O i. The absorption component of the 63 μm line is reproduced with an O i/OH abundance ratio of 103, close to the value expected for translucent clouds. We first ignore the possible contribution by the nucleus to the emission feature of the [O i] 63 μm line. The resulting line shape is compared with the observed profile in Fig. 11a (green curve), and shows that (i) the velocity field and density profile adopted for OH roughly match the absorption in the [O i] 63 μm line, though the modeled absorption is somewhat overestimated at low velocities; (ii) in our spherically symmetric model, the emission feature is formed in the inflowing envelope through scattering of 63 μm photons, overestimating by 70% the flux of the observed spectral feature (325 Jy kms-1 between −200 and 0 kms-1). Our model for O i thus supports an inflow with wide opening angle, but with the amount of gas inflowing on the lateral sides being lower than in front of it.

The above solution for the blueshifted emission in the [O i] 63 μm line, i.e. that the feature is formed in the inflowing envelope, is, however, not unique, because of the possible contribution by the nucleus to this feature. Indeed, the [O i] 145 μm line, though weak and too narrow as compared with the instrumental spectral resolution, is detected in the PACS spectrum (Fig. 11b), and the line should be formed in the nuclear region. Here we attempt to address the conditions under which the emission in the [O i] 63 μm line may be also formed in the nuclear region. We have then calculated the expected nuclear emission in the [O i] 63 μm line as a function of the assumed nuclear gas density. In all models, the nuclear O i column is fixed to match the observed [O i] 145 μm line. We have used a “mixed” approach as the continuum opacity at 145 μm is moderate and the inner regions of the nucleus can contribute to the [O i] 145 μm line. We have adopted the dust model for the Cwarm component with Td = 110 K, and the gas at Tg = 150 K. Collisional rates between O i and H2 are taken from Jaquet et al. (1992). Results for the blueshifted (i.e. integrated between −200 and 0 kms-1) emission are compared with the observed value in Fig. 20. At the relatively low density of 105 cm-3, the excitation of the [O i] 63 μm line is below Td, and the line is predicted in weak absorption. However, the absorption changes to emission for higher densities, and the blueshifted integrated flux attains a value comparable to the observed one at nH2 ≈ 3 × 106 cm-3. Thus a significant contribution by the compact nucleus to the [O i] 63 μm emission feature is not ruled out from the point of view of the radiative transfer. For nH2 = 3 × 106 cm-3, the “mixed” O i column is NO0 ≈ 1.5 × 1020 cm-2, and χO0 ~ 3 × 10-5.

The case of the [O i] 63 μm emission appears then somewhat ambiguous on the basis of these radiative transfer models. Still, the OH models and the requirement of an extended region to account for the emission in the OH 163 μm doublet, favor Cextended as the main contributor to the [O i] 63 μm emission feature, but the relative contribution by the nucleus is hard to constrain. We further discuss this issue in Sect. 4.4.

thumbnail Fig. 20

Modeled [O i] 63 μm emission above the continuum, integrated over blueshifted velocities (i.e. between −200 and 0 kms-1 with respect to the systemic velocity), for the nucleus of NGC 4418 as a function of the H2 density. The dashed line indicates the observed value.

Open with DEXTER

3.7. The continuum of NGC 4418

Our continuum fit to NGC 4418, shown in Fig. 1a and with parameters listed in Table 1, is composed of: (i) a hot component (Chot, yellow curve) with Td = 350 K and an effective diameter of d ≈ 5 pc, attenuated by AV = 70 mag. It accounts for the mid-IR emission including the silicate 10 μm absorption; (ii) the nuclear core (Ccore, blue curve) with Td = 150 K and d = 20 pc, attenuated by AV = 81 mag. It is required to explain the high-lying molecular absorption observed in H2O, HCN, OH, and NH3; (iii) the warm component (Cwarm, light-blue curve) with Td = 110 K and d = 31 pc, attenuated by AV = 65 mag. The parameters of this component are somewhat uncertain and could include regions with lower Td ~ 85 K. Cwarm mainly accounts for some mid-excitation lines of H2O and OH; (iv) the extended component (Cextended, solid green curve) with non-uniform Td and d ≈ 200 pc, which mainly accounts for the redshifted OH absorption and for the emission in the OH 163 μm doublet. The emission from the above components still underestimates the observed continuum at λ ≳ 100 μm, and therefore we have added a more extended, cold component with d = 650 pc and Td = 50−25 K (dashed-green curve in Fig. 1a). This cold component, however, has no obvious molecular counterpart at far-IR wavelengths and is included in order to obtain a reasonable fit to the SED at long wavelengths, though the total emission (red curve in Fig. 1a) still underestimates the (sub)millimeter emission. Possible causes are the wavelength dependence of the dust absorption coefficient, and the uncertain dust opacity and size of the nuclear core at (sub)millimeter wavelengths.

The unattenuated luminosities of the hot, core, and warm components are 8.5 × 1010 L (i.e. 85% of the bolometric luminosity),  ~7 × 1010 L, and 6 × 1010 L, respectively. The sum of the three exceeds the bolometric luminosity, implying that, unless there are important geometrical effects that make us underestimate the true galaxy luminosity, most emission of the core and warm components is not intrinsically generated, but largely reemission due to heating from inside. Nevertheless, clear departures from spherical symmetry are inferred from our modeling, as e.g. the extinction affecting Chot is much lower than the extinction through Ccore , so the latter component can not be interpreted as a homogeneous shell surrounding Chot .

4. Discussion

4.1. Model uncertainties

Our schematic models adopt uniform densities and uniform dust temperatures for the nuclear components, which is clearly a rough approach as it ignores their inner structures and clumpiness. The use of spherical symmetry is also certainly inaccurate, especially in NGC 4418 where a linear structure for the nuclear region is observed in the near-IR (Evans et al. 2003). Thus our models should be considered equivalent spherical/uniform approaches to complex structures.

Toward the nuclear region of Arp 220 and NGC 4418, our column density estimates are dust-limited, in the sense that the line absorption traces the outermost regions where τd < 1. In this far-IR photosphere, we estimate that our derived columns per unit of τ50 are probably accurate to within 50% for a given Td. Conversion to abundances is more uncertain. We have normalized the inferred column densities to a total H column of NH,norm = 4 × 1023 cm-2. However, the above H column depends on the gas-to-dust ratio as well as on the assumed properties of dust grains. Furthermore, the HCN absorption is observed at wavelengths longer than most H2O and OH lines, and thus traces higher H columns than the latter species. The molecular abundances are thus reliable only to within a factor of 2−3. Nevertheless, abundance ratios between different molecular species are based on the same approach, and are therefore more accurate.

4.2. The nuclear regions of NGC 4418 and Arp 220

4.2.1. A hot core chemistry

The far-IR Herschel/PACS observations of NGC 4418 and Arp 220 reveal extreme excitation in the key molecular species H2O, OH, HCN, and NH3. The lines are observed mostly in absorption against the continuum, and our analysis indicates that very high column densities and abundances of these molecules surround (and are associated with) compact, far-IR thick, luminous continuum components.

One of the most intringuing results of our analysis is the high abundance derived for HCN, (1−2) × 10-6 in both Arp 220 and NGC 4418. At millimeter wavelengths, HCN has been found to be enhanced over HCO+ in ULIRGs (Graciá-Carpio et al. 2006) and in Seyferts over starbursts (Imanishi et al. 2007), and this has been proposed to be a diagnostic of AGN activity. Costagliola et al. (2011), however, argue that for high densities and columns HCO+ is expected to be stronger than HCN in XDRs (Meijerink et al. 2007). Models by Meijerink & Spaans (2005) predict HCN abundances and cumulative column densities in XDRs that fall by far too short to explain the values we derive in both NGC 4418 and Arp 220. Likewise, the peak HCN abundances predicted by Lepp & Dalgarno (1996) for XDRs, a   few × 10-8, are about two orders of magnitude lower than the values we derive.

The high HCN abundances in both sources are very likely tracing a hot-core like chemistry, as has been proposed by previous authors, i.e. by Lahuis et al. (2007) from the HCN 14 μm absorption in these and other (U)LIRGs, Martín et al. (2011) for Arp 220 based on multispecies observations at 1.3 mm, and Aalto et al. (2007a) and Costagliola & Aalto (2010) to explain the luminous HC3N emission in NGC 4418. HCN is well known to be enhanced in Galactic hot cores, as in the Orion hot core where the abundance is  ~10-6 (Schilke et al. 1992), about two orders of magnitude higher than in cold molecular clouds. Time-dependent chemical models for hot cores by Rodgers & Charnley (2001) predict a linear increase of the HCN abundance with time for Tg = 300 K, attaining values of 10-5 relative to H2 after  ~106 yr of chemical evolution. The authors describe this model result as “pathological” as such high abundances are not found in Galactic hot cores, and suggest that these regions do not survive for more than a million years. However, extragalactic nuclear regions may survive long enough to develop such high HCN abundances.

The proposed evolved hot core chemistry finds support in the HCN/NH3 ratio of 2–5 estimated in both Arp 220 and NGC 4418. After  ≳106 yr of chemical evolution, the models by Rodgers & Charnley (2001) predict that a similar value of the ratio can be reached provided that the gas temperature is high enough, though HCN is predicted to continue increasing with time. The ratio found by Rodgers & Charnley (2001) at such late stages of hot core evolution is independent of whether NH3 was injected into the gas phase from mantles or formed directly in the gas phase.

The hot-core scenario is also quite consistent with the warm dust temperatures derived in the nuclear regions of NGC 4418 and Arp 220, and with the derived high H2O abundances. In the Orion hot core, Melnick et al. (2010) derive H2O abundances of  ~10-5 from Herschel/HIFI observations. Theoretically, an “undepleted chemistry” where grain mantles are evaporated is expected to increase the gas-phase H2O abundance up to values  >10-5 (e.g. Charnley 1997), i.e. orders of magnitude relative to the values inferred in cold molecular clouds and PDRs (Melnick & Bergin 2005), where H2O ice is expected to be efficiently formed upon hydrogenation of atomic oxygen depleted on mantles in cold environments (Bergin et al. 2000; Hollenbach et al. 2009). If the gas attains sufficiently high temperatures (Tg ~ 300 K), H2O will be further enhanced in the gas phase via neutral-neutral reactions (e.g. Elitzur & de Jong 1978).

The current models underpredict the low-J lines of HCN observed by Aalto et al. (2007a) in NGC 4418, who derived densities of  ~105 cm-3; we derive much higher densities,  ~3 × 106 cm-3, for the high-J lines. Likewise, our models for the high-J lines in Arp 220 underpredict the J = 6 → 5 measurement of Rangwala et al. (2011). It is plausible that these sets of lines probe different components, with the higher-J lines tracing very dense gas close to the regions of high far-IR emission with temperatures of Td > 100 K, while the low-J lines trace a surrounding or interclump component with more moderate densities. No lines of HNC are found in the PACS spectra of NGC 4418 or Arp 220, while at millimeter wavelengths HNC is overluminous relative to HCN (Aalto et al. 2007b).

The high OH abundances in both NGC 4418 and Arp 220 do not easily fit into the hot core scenario. Observationally, OH observations of Orion BN/KL by Melnick et al. (1990) indicate that the emission/absorption is mostly produced by the shocked gas associated with the outflow, with little contribution, if any, by the hot core. Recently, Wampfler et al. (2011) derived a very low OH/H2O ~ 10-3 abundance ratio in the warm (T > 100 K), quiescent component of the Galactic high-mass star-forming region W3 IRS 5. Models by Charnley (1997) indeed predict a very low OH/H2O ratio in hot cores as most oxygen at high T is expected to be chanelled into H2O.

4.2.2. X- and cosmic rays

Another component, with chemistry dominated by photoprocesses and/or cosmic rays, is thus required to account for the OH enhancement in the nuclear region of both galaxies. Multitransition observations of OH in the Orion bar and chemical modeling by Goicoechea et al. (2011) reveal a dense/warm PDR component with an OH abundance of a   few × 10-6. However, the OH enhancement is expected to be restricted to the external AV ≲ 1 layers where OH/H2O > 1 due to H2O photodissociation by the FUV field (Goicoechea et al. 2011). In NGC 4418 and Arp 220, we require widespread OH to account for the observations and, if PDRs alone are invoked, the surfaces of the nuclear regions in both sources (with typical sizes of 40 and 100 pc) would have to be covered with such high-density, low-AV PDRs with a covering factor of  ≳0.5 and, for typical individual PDR columns of NOH ~ a   few × 1015 cm-2 (Goicoechea et al. 2011),  ≳100 of such regions are needed along the line of sight. The impact of FUV photons on large amounts of gas can be enhanced under special geometries, as in the walls of cavities excarvated by outflows (Bruderer et al. 2010); OH is boosted in outflows where H2O is photodissociated by strong FUV fields (Melnick et al. 1990; Spinoglio et al. 2000). However, again a large number of overlapping outflows would be required across columns of several × 1023 cm-2. Furthermore, the outflow signatures in Arp 220 are only observed in lines of moderate excitation, and NGC 4418 does not show any hint of outflowing gas, indicating that high-velocity (as compared with Galactic standards) outflows are not responsible for the observed chemistry. Nevertheless, the presence of Orion-like outflows with velocities of  ~30 kms-1 is not ruled out.

A promising scenario to explain the high OH columns and abundances is through the effect of cosmic/X-rays, which penetrate in high-AV regions much more deeply than UV photons. The OH/H2O ratio as predicted by XDR and CR models of high density molecular regions is  >1, and χOH easily exceeds 10-6 (Maloney et al. 1996; Meijerink et al. 2011). Large amounts of OH can be formed through dissociative recombination of H3O+, which in turn is formed through the sequence O+ → OH+ → H2O+ → H3O+ (Maloney et al. 1996); as we will report in a forthcoming paper, these molecular ions are indeed detected in absorption from excited levels in both sources. However, both NGC 4418 and Arp 220 have strong X-ray deficits relative to the FIR emission (Maiolino et al. 2003; Iwasawa et al. 2005). From the X-ray flux reported by Maiolino et al. (2003) from NGC 4418, FX(0.3−8   keV) = 1.5 erg   s-1 cm-2, and assuming a point-like X-ray source located at the center of our Ccore component, the X-ray flux at the surface of Ccore (r ~ 10 pc) is  ~0.1 erg   s-1 cm-2; a similar estimate for Arp 220 with the fluxes reported by Iwasawa et al. (2001) yields an X-ray (0.5−10 keV) flux at the surface of the western nucleus (r ~ 50 pc) of  ~0.5 erg   s-1 cm-2. Since the lines observed with PACS are formed around far-IR thick components with NH ≳ 5 × 1024 cm-2, any putative strong and buried X-ray source would be severely extincted, and the above distance-corrected estimates are probably lower limits due to foreground absorption but still relevant within a factor of 3. Conversion to ionization rates (ζ) and energy deposition rates per unit of hydrogen density (HX/n) is very sensitive to the adopted X-ray spectrum (Wardle et al. 1999); adopting ζ   (s-1) ~ 2 × 10-13FX   (erg   s-1 cm-2), we infer ζ ~ (0.2−1) × 10-13 s-1 and HX/n ≲ (1−5) × 10-30 erg cm3 s-1 (Maloney et al. 1996) for NGC 4418-Arp 220 in the far-IR nuclear photosphere traced by the OH. It is unclear if such low energy deposition rates per unit density can produce the high nuclear OH abundance we measure at least in NGC 4418, where very high densities are found. Nevertheless, cosmic rays are an alternative, and both X- and cosmic rays could efficiently generate OH from a high reservoir of H2O that is formed by other means (i.e. by a hot core chemistry or by low-velocity outflows), through photodissociation of H2O by internally generated UV photons (Gredel et al. 1989; Wardle et al. 1999) and/or reactions of H2O with He+, H+, , and to form OH directly or via the H2O+ and H3O+ intermediates; this high H2O reservoir is evident in both sources. Detailed chemical modeling is required to differentiate between these, and other alternatives; while H2O may be expected to be abundant in regions where HCN-NH3 or OH are enhanced, it remains to be determined whether the HCN-NH3 and the OH can be cospatial.

4.2.3. Sources of luminosity

The presence of high columns of HCN and H2O and the implied hot core chemistry in the nuclear regions neither supports nor rules out the presence of a deeply buried AGN. It could simply indicate that, regardless of the nature of the intrinsic luminosity source(s), frequency degradation in a very thick medium (as UV radiation is absorbed by dust and re-emitted at longer wavelengths by sampling progressively colder media), makes far-IR photons more relevant for the chemistry of large fractions of the nuclear gas (through dust heating and mantle evaporation) than the quickly absorbed UV photons.

Taking the Galactic Orion hot core as a template, its size (≈10′′ diameter, or 0.02 pc at 414 pc; e.g. Wilson et al. 2000) is  ~900 and  ≳4000 times smaller (and the square of these numbers in surface) than those estimated for the Ccore and Cwest components of NGC 4418 and Arp 220, respectively. If the HCN absorption is interpreted in terms of hot cores around young stars, then  ~8 × 104 and  ~1.5 × 107 cores are inferred, respectively. Likewise, if a typical hot core harbors a luminous star with 105 L,  ~7 × 105 and  ~7 × 106 young stars are inferred, respectively. This interpretation, however, has the drawback that Galactic hot cores are short lived, with a time scale of order 105 yr (Wilner et al. 2001). Even if the high pressure in the nuclear regions (Spoon et al. 2005) allows the hot cores to survive longer, a synchronized birth of  ~107 hot cores in  <1 Myr over spatial scales of  ≳80 pc, involving a propagation velocity of stellar formation of  >100 kms-1, seems unlikely. The observed absorption is rather interpreted in terms of large hot-core like structures (disk/torus/envelope) not necessarily associated with the very first stages of star formation.

The inferred hot core chemistry, complemented by the strong effects of X- and/or cosmic rays, is found to be physically associated with warm and luminous sources of far-IR emission. The surface brightnesses of the nuclear regions in both sources are very high,  ~7 × 1013 L/kpc2 over 20 pc for the Ccore component of NGC 4418, and (0.9−4) × 1013 L/kpc2 over 150−80 pc for the western nucleus of Arp 220. In the latter source, these are lower limits if the dust is heated from inside (Downes & Eckart 2007). Sa8 indicates that the luminosity from the Arp 220 western nucleus could be generated by 200−1000 super-star clusters. An important contribution from star formation to the luminosity of the western nucleus is indeed evident (e.g., Smith et al. 1998; Lonsdale et al. 2006), and the apparent lack of strong X-ray emission from X-ray binaries associated with star formation (Lonsdale et al. 2006) can be interpreted in terms of extreme absorption column densities (Sa08, this work).

On the other hand, the broad 1.6 GHz OH megamaser emission terminal velocity of 800 kms-1 in Arp 220 (Baan et al. 1989), comparable to those of OH in the far-IR absorption lines of AGN driven ULIRGs (Fischer et al. 2010; Sturm et al. 2011), is much higher than the OH velocities observed in the far-IR spectrum of Arp 220. We interpret this as evidence for both far-IR extinction of the high velocity gas accelerated in the innermost nuclear region and deceleration of the outflowing gas with increasing involved columns. A very compact source for the outflowing gas would also support the presence of an energetically significant, strongly buried AGN that would also explain the somewhat different distributions traced by the compact centimeter-wave sources and the submillimeter continuum emission (Sa08). In NGC 4418, the extreme HCN and H2O excitation on a scale of 20 pc in diameter, the large fraction of luminosity coming from such a compact region, and the indications of internal heating, favor the scenario of a far-IR thick disk/torus/envelope surrounding a nascent AGN, as has been proposed in previous works (e.g. Spoon et al. 2001), or a hyper-star cluster  ~70 times more luminous than the super-star cluster in NGC 5253 (Turner et al. 2000).

In NGC 4418, the outward acceleration due to radiation pressure can be described by  cm s-2, where κ ~ 2−5 cm2 g-1 is the Rosseland mean opacity at 100−150 K (Thompson et al. 2005) and σ is the Stefan-Boltzmann constant. We find that arad is similar to the gravitational acceleration, agrav = GM/R2 ~ (2.5−5) × 10-6 cm s-2, where M = (1.6−3.2) × 107 M (for τ200 ~ 1−2) and R = 10 pc have been estimated (Table 1). If the stellar mass is not much higher than the gas mass in the Ccore component, the above comparison suggests that radiation pressure plays an important role in supporting the nuclear structure against self-gravity in an Eddington-like limit (Scoville 2004). In Arp 220 (Cwest) radiation pressure appears less important due to the higher M/R2 and lower4Td. Nevertheless, the disk in the western nucleus may be inclined such that our view of the warmer dust is obscured in the far-IR and we have a “rotationally-supported” edge-on view of the structure.

4.3. The extended component of Arp 220

The extended and halo components (Cextended and Chalo) in Arp 220 reveal a chemistry very different from that found in the nuclear region. The H2O abundance drops to the relatively low value of  ~5 × 10-8, typical of Galactic PDR regions, and the OH/H2O ratio reverses the nuclear value becoming  >1, as in Galactic translucent clouds (e.g. Neufeld et al. 2002). Our estimated OH abundances of several    × 10-7 are in general agreement with those derived by Polehampton et al. (2005) in the intervining translucent clouds toward Sgr B2. On the other hand, the outflow signatures observed in the OH 163 and 79 μm doublet, in the [O i] 63 and 145 μm lines (Fig. 11), and in the H2O 303 → 212 line (Fig. 4p), appear to indicate that the outflow is spatially extended on the back side of the nuclei.

The extended component of Arp 220 resembles the translucent Galactic medium, though with a higher ionization rate as derived from the absorption in lines of molecular ions, as we will report in a forthcoming paper.

4.4. Inflowing gas in NGC 4418?

The intringuing redshifted absorption in the ground-state lines of OH and O i, and the blueshifted emission in the [O i] 63 μm line, caused us to consider in Sect. 3.6 the possibility that an extended envelope around the nucleus of NGC 4418 is inflowing onto the nuclear region, thus feeding the nucleus and providing an external ram pressure that may further confine the nuclear gas. Krug et al. (2010) have reported the detection of inflows in Seyferts based on absorption measurements of the Na I doublet at optical wavelengths.

The morphology of the region as seen at near-IR wavelengths (Evans et al. 2003) and the absorption depths indeed suggest that the redshifted absorbing component is more than a single cloud in the galaxy disk with an anomalous velocity; the redshifted absorption affects the whole column of gas external to the nuclear region (~1023 cm-2), as otherwise the ground-state OH absorption at the systemic velocity would be deeper than observed. Thus the OH and O i absorptions suggest bulk inward gas motions.

If spherical symmetry is assumed for this component, a mass inflow rate of 12 M yr-1 is obtained at the surface of the nucleus, based on the model shown in Fig. 11. Spherical symmetry is consistent with the OH line profiles, in particular with the emission in the OH 163 μm doublet. However, the spherical approach overpredicts the [O i] 63 μm blueshifted emission (Fig. 11a). Furthermore, pure radiative transfer cannot rule out that a significant fraction of the [O i] 63 μm emission feature is formed in the nucleus, rather than in the envelope, if the high density component (several  × 106 cm-3, as derived from OH, H2O, and HCN) is sufficiently rich in atomic oxygen (a few  × 10-5). Observational surveys of the [O i] 63 μm line in Galactic hot cores may help further evaluate the latter possibility, but pure hot core chemistry predicts O i to have an abundance of only  ≲10-6 (Charnley 1997). Guided by these models, we do not expect much contribution from dense hot cores to the [O i] 63 μm emission feature. However, the high-lying OH lines are tracing a region where photodissociation of H2O is probably taking place (Sect. 4.2.2), so OH photodissociation could produce significant columns of dense O i. Still, the C-shock models by Wardle et al. (1999) with enhanced H2O photodissociation due to X- or cosmic rays produce little O i, with OH/Oi > 100, but models for XDRs and CRs by Meijerink & Spaans (2005) and Meijerink et al. (2011) predict large amounts of oxygen in atomic form, though they use densities significantly lower than 3 × 106 cm-3.

In summary, our observations and analysis favor an inflow in NGC 4418, though the geometry and inflow rate of the inflowing gas are rather uncertain. If the gas were inflowing only from the front side of the nucleus with an opening angle of, say, 90°, the mass inflow rate would be of only  ≈2 M yr-1. Kawara et al. (1990) first suggested a possible inflow in NGC 4418 based on the narrow CO J = 1−0 line. The interaction with the nearby galaxy VV 655 could have trigered the inflow via a close (perigalactic), non-merging encounter  ~108 yr ago for an average velocity difference of  ~200 kms-1 in the plane of the sky, and models (Noguchi 1988; Iono et al. 2004) predict that the inflow triggered by interacting galaxies may last that long. Alternatively, a non-axisymmetric gravitational potential as in barred galaxies can fuel the nuclear region with large amounts of gas (Friedli & Benz 1993), which is supported by the peculiar morphology of NGC 4418.

4.5. Evolutionary effects

In spite of the very different global parameters (mass, luminosity, size, and merger versus non-merger class) of Arp 220 and NGC 4418, their nuclear regions share common properties. They contain warm dust, very excited molecular absorption, and high surface column densities of H2O, OH, HCN, and NH3. There are, however, significant differences in the properties revealed by their far-infrared spectra that probably stem from and further probe their individual evolutionary histories. The smaller, less massive, and less luminous NGC 4418 shows warmer dust and higher molecular excitation than Arp 220. Perhaps the most suggestive spectroscopic differences between the two sources are their inferred kinematics and 16O to 18O ratios: while NGC 4418 appears to have a ratio similar (to within a factor of 2) to that seen in the local solar neighborhood, the ratio in Arp 220 is significantly lower.

The outflow signatures in Arp 220, which range from 800 kms-1 in its OH megamaser radio emission to 100 kms-1 in submillimeter transitions of HCO+ and CO and up to 500 kms-1 in some moderate excitation submillimeter and far-IR emission lines, suggest that injection of enriched gas from current and previous stellar processing has occurred. On the other hand, mechanical feedback has not yet occurred in NGC 4418, which may still be in a stage of inflow. Analysis of the O and the 18OH lines indicate 16O-to-18O ratios of 70−130 in Arp 220, reminiscent of the low ratio of  ~40 derived for Mrk 231 (Fischer et al. 2010). The strong absorption in the red Λ−component of the ground-state 18OH 120 μm doublet (Fig. 8c) indicates that the 18O enhancement is not restricted to the nuclear region, but also applies to the Cextended and Chalo components. At millimeter wavelengths, Martín et al. (2011) derive a 16O-to-18O ratio of  >80 in Arp 220, while Harrison et al. (1999) estimated a ratio of  ~150 in NGC 253. In NGC 4418, however, we find that the 16O-to-18O ratio is  ≳250, with a best fit value of  ~500. While 16O is a primary product of stellar nucleosynthesis, 18O is a secondary one, and this ratio decreases with increasing star forming generations (Prantzos et al. 1996). Since galactic outflows are dominated by entrained ISM gas (Veilleux et al. 2005), the low ratios found in Arp 220 and Mrk 231 probably trace stellar processing throughout the merger lifetime. The apparent ISM inflow, lack of outflowing gas, and higher 16O-to-18O ratios in NGC 4418 appear to indicate an earlier evolutionary phase in the lifetime of the central AGN or nuclear starburst.

We tentatively suggest that the nuclei of Arp 220 and NGC 4418 are in an earlier evolutionary phase than the QSO driven ULIRG Mrk 231 (e.g. Veilleux et al. 2009). In this source, high velocity (~1000 kms-1) feedback (Fischer et al. 2010; Sturm et al. 2011; Rupke & Veilleux 2011) in the form of a massive molecular/atomic outflow has more strongly disrupted its nuclear interstellar medium, as evidenced by the lower dust opacity of its warm component (González-Alfonso et al. 2010) compared with the nuclear regions of NGC 4418 and Arp 220. The lack of measurable HCN emission/absorption in its Herschel/SPIRE spectrum (van der Werf et al. 2010), suggests that exposure to UV and X-ray emission from its AGN appears to have at least partially shut down the hot core nitrogen chemistry5. Further, analysis of Herschel/SPIRE observations of H2O emission in Mrk 231 (González-Alfonso et al. 2010) suggest an H2O abundance of  ~10-6, significantly lower than in NGC 4418 and Arp 220, but observations of the same lines would help verify this issue. Lastly, the stronger stellar processing suggested by the increased enhancement of 18OH in Mrk 231 may indicate an older age of its merger induced star-formation than in Arp 220. This evolutionary scenario is consistent with the fact that in the mid-IR Mrk 231 looks AGN-like, while Arp 220 and NGC 4418 are still dominated by deep silicate absorption indicative of a buried, earlier evolutionary stage (Spoon et al. 2007).

5. Conclusions

The salient observational results presented here on NGC 4418 and Arp 220 are:

  • Full range PACS spectroscopy of NGC 4418 andArp 220 reveals high excitation inH2O  OH, HCN, and NH3. In NGC 4418, absorption lines were detected with Elower > 800 K (H2O), 600 K (OH), 1075  K (HCN), and 600 K (NH3), while in Arp 220 the excitation is somewhat lower with Elower > 600 K (H2O), 600 K (OH), 650 K (HCN), and 400 K (NH3).

  • More specifically, NGC 4418 shows 38 H2O lines in absorption, and with marginal detections of even higher-lying lines, while Arp 220 shows absorption in 28 H2O lines. All strong OH lines within the PACS domain are detected in both sources. In NGC 4418, at least 6 HCN rotational lines, from J = 18 → 17 to J = 23 → 22 with Elower from 650 to 1075 K are detected in absorption, while the J = 18 → 17 and possibly the J = 21 → 20 lines are detected in Arp 220. Rotational NH3 lines arising from Elower up to  ~600 K are detected in NGC 4418, while Arp 220 shows NH3 lines up to  ~400 K. Molecular absorption equivalent widths from high-lying levels are higher in NGC 4418 than in Arp 220.

  • Up to 10 lines of and 3 18OH Λ-doublets are detected in Arp 220, while 6 O lines are detected in NGC 4418, where only a marginal detection of 18OH is found. The lines of the rare isotopologues are thus significantly stronger in Arp 220 than in NGC 4418.

  • Lines with moderate excitation in Arp 220 show indications of outflowing gas, specifically in the redshifted emission features observed in lines of H2O, OH, and O i. The absorption in more excited molecular lines peak at central velocities relative to our adopted redshift of 0.0181, though a contribution to the absorption by outflowing gas with velocities up to  ~150 kms-1 is not ruled out. In NGC 4418, however, the molecular lines do not show any indication of outflowing gas, and the absorption in the low-lying OH doublets as well as in the [O i] 63 μm line are redshifted.

We have analyzed the observed absorption-dominated line profiles together with the far-IR continuum, using multicomponent radiative transfer modeling with a minimum number of parameterized components. The conclusions stemming from the analysis are:

  • Extremely high H2O column densities per unit ofcontinuum optical depth at 50 μm (τ50),  ≳3 × 1018 cm-2, and OH/H2O and HCN/H2O ratios of  ≈0.5 and 0.1−0.3, respectively, are derived in the nuclear region of NGC 4418, along with dust temperatures  ~150 K in the very inner nuclear core. In the nuclear region of Arp 220, Td is in the range 90−130 K, implying H2O columns of (6−0.8) × 1018 cm-2 per τ50 and OH/H2O and HCN/H2O ratios similar to those of NGC 4418. The HCN/NH3 ratios are 2−5 in both sources. The continuum emission in these nuclear regions is optically thick at far-IR wavelengths in both galaxies.

  • The nuclear molecular abundances relative to H are estimated to be high, with χH2O ~ 10-5 in NGC 4418 and (0.2−1.5) × 10-5 in Arp 220.

  • The absorption by the 18O isotopologues indicate a strong enhancement of 18O in Arp 220, with a 16O-to-18O ratio of 70−130, while the ratio in NGC 4418 is significantly higher,  ≳250 with a best fit value of  ~500. The apparent 18O enhancement in Arp 220 probably traces stellar processing throughout the merger lifetime.

  • We also require an extended component to produce the absorption in the low-lying lines of OH and H2O; the fractional H2O abundance in Arp 220 drops in this region to  ~5 × 10-8 and the OH/H2O is reversed relative to the nuclear region to 2.5−10. In the envelope of NGC 4418, we find χOH = a   few × 10-7.

  • The extreme H2O, HCN, and NH3 columns and abundances in the nuclear region of both galaxies, together with the high dust temperatures, are interpreted in terms of a hot core chemistry, where grain mantles are evaporated. The high OH columns and abundances require in addition photoprocesses due to X-rays and/or enhancement by high cosmic ray fluxes. H2O and OH may also be boosted in low-velocity shocks with enhanced H2O photodissociation.

  • Very thick (NH ≳ 1025 cm-2) nuclear media are responsible for the frequency degradation of the intrinsic luminosity in both sources, as high energy photons are absorbed and re-emitted at far-IR wavelengths sampling progressively colder media. Such conditions, and not necessarily a starburst-dominated luminosity, may also be responsible for the hot core chemistry in the nuclear regions. The high surface brightnesses,  ≈6 × 1013 and (0.9−4) × 1013 L/kpc2 on spatial scales of  ~20 and 150−80 pc in NGC 4418 and Arp 220, respectively, indicate either strongly buried AGNs or hyper star clusters. In NGC 4418, radiation pressure is probably responsible for the support of the nuclear structure against self-gravity.

  • Both the outflowing gas and the low 16O-to-18O ratio in Arp 220 suggest a stage more evolved than in NGC 4418, but less evolved than the QSO driven ULIRG Mrk 231.

  • In NGC 4418, the redshifted absorption observed in the low-lying OH and [O i] lines is best explained in terms of an envelope inflowing onto the nuclear region, but the geometry of this inflow is rather uncertain; we estimate the mass inflow rate to be  ≲12 M yr-1.


1

PACS has 5 × 5 spatial pixels (spaxels), each with 9′′ × 9′′ resolution and covering a field of 47′′ × 47′′ (Poglitsch et al. 2010).

2

Since the neighboring spaxel has a signal much lower than the central one, the subsequent final calibration leaves the factor nearly unchanged.

3

Even if the mid-IR arises from the two nuclei, the continuum is simulated with just a single component with an effective diameter as given in Table 1.

4

Vertical radiation pressure support in optically thick nuclear disks imply (e.g. Thompson et al. 2005), where Teff is the effective dust temperature and Σg is the gas column. Comparison of NGC 4418 (Ccore) and Arp 220 (Cwest) indicates, however, lower Teff and higher Σg in Arp 220.

5

But not completely: during the final revision of this manuscript, Aalto et al. (2012) reported the detection of bright HC3N (10–9) in the nuclear region of Mrk 231.

Acknowledgments

We thank David S. N. Rupke for deriving the redshift of NGC 4418 from SDSS, Kazushi Sakamoto for providing us with the spectra of the HCO+ (3–2) and (4–3) lines in the nuclei of Arp 220, and the referee Christian Henkel for many useful indications and comments that much improved the manuscript. PACS has been developed by a consortium of institutes led by MPE (Germany) and including UVIE (Austria); KU Leuven, CSL, IMEC (Belgium); CEA, LAM (France); MPIA (Germany); INAFIFSI/OAA/OAP/OAT, LENS, SISSA (Italy); IAC (Spain). This development has been supported by the funding agencies BMVIT (Austria), ESA-PRODEX (Belgium), CEA/CNES (France), DLR (Germany), ASI/INAF (Italy), and CICYT/MCYT (Spain). E.G.-A. thanks the support by the Spanish Ministerio de Ciencia e Innovación under project AYA2010-21697-C05-01, and is a Research Associate at the Harvard-Smithsonian Center for Astrophysics. Basic research in IR astronomy at NRL is funded by the US ONR; J.F. also acknowledges support from the NHSC. S.V. thanks NASA for partial support of this research via Research Support Agreement RSA 1427277. He also acknowledges support from a Senior NPP Award from NASA and thanks his host institution, the Goddard Space Flight Center. This research has made use of NASA’s Astrophysics Data System (ADS) and of GILDAS software (http://www.iram.fr/IRAMFR/GILDAS)

References

All Tables

Table 1

Parameters of the continuum models.

Table 2

Derived H2O column densities and abundances, column density ratios, and covering factors of the continua.

Table 3

H2O line equivalent widths, fluxes, and velocity shifts.

Table 4

[O i] and OH line equivalent widths, fluxes, and velocity shifts.

Table 5

HCN line equivalent widths, fluxes, and velocity shifts.

Table 6

NH3 line equivalent widths and fluxes for relatively isolated spectral features.

Table 7

O line equivalent widths and fluxes.

Table 8

18OH line equivalent widths and fluxes.

All Figures

thumbnail Fig. 1

Spectral energy distribution of NGC 4418 and Arp 220 from mid-IR to millimeter wavelengths. Both the Herschel/PACS and the Spitzer/IRS spectra are shown. Data points at (sub)millimeter wavelengths for NGC 4418 are from (Yang & Phillips 2007, 350 μm), (Roche & Chandler 1993, 450, 800, and 1100 μm), (Dunne et al. 2000; Lisenfeld et al. 2000, 850 μm), and (Albrecht et al. 2007, 1300 μm). For Arp 220, they are taken from (Eales et al. 1989, 450 μm); (Rigopoulou et al. 1996, 350, 800, and 1100 μm); and (Sakamoto et al. 1999, 2008, 1300 μm). Models discussed in Sect. 3 are included. For NGC 4418, the yellow, blue, light-blue, and solid-green curves show the models for the hot, core, warm, and extended components, respectively, with parameters given in Table 1. An additional cold component, shown with the dashed-green curve, is included to better fit the SED at long wavelengths. For Arp 220, the yellow, blue, light-blue, and solid-green curves show the models for the hot, western nucleus, eastern nucleus, and extended components, respectively (Table 1).

Open with DEXTER
In the text
thumbnail Fig. 2

Energy level diagrams of OH and H2O (ortho and para). Red arrows indicate lines detected in both sources; blue arrows denote lines detected only in NGC 4418, green-dashed arrows mark transitions contaminated (blended) by lines of other species, and blue-dashed arrows denote marginal detections in NGC 4418. The small numbers in the OH diagram indicate rounded wavelengths in μm.

Open with DEXTER
In the text
thumbnail Fig. 3

H2O lines in the blue and green bands (λ < 100 μm) detected in NGC 4418 (upper spectra in each panel) and Arp 220 (lower spectra). The spectra, histograms with black solid lines, have been scaled by a factor of 0.5 in panels k) and p). Vertical dotted lines indicate the rest wavelengths of the lines relative to z = 0.00705 and 0.0181 for NGC 4418 and Arp 220, respectively. Model results are also shown. For NGC 4418, the blue and light-blue curves show the models for the Ccore and Cwarm components, respectively; for Arp 220, the blue and light-blue lines show the estimated contributions from Cwest and Ceast , and the green curve (in panel k) shows the contribution by Cextended . Red is total. Model parameters are given in Tables 1 and 2.

Open with DEXTER
In the text
thumbnail Fig. 4

H2O lines in the red band (λ > 100 μm) detected in NGC 4418 (upper spectra in each panel) and Arp 220 (lower spectra). The spectra have been scaled by a factor of 2 in panels a), k) and l), and by 3 in panel c). Model results are also shown. Color codes for the contributions of the different components are given in Fig. 3, with the yellow curve in panel q) showing the additional contribution by the halo component (Chalo) in front of the nuclear region of Arp 220.

Open with DEXTER
In the text
thumbnail Fig. 5

Marginally detected H2O lines in NGC 4418 (upper spectra in each panel); the corresponding lines from Arp 220 are also shown (lower spectra). Model predictions for these lines (for Ccore in NGC 4418 and for Cwest in Arp 220) are also shown (red curves).

Open with DEXTER
In the text
thumbnail Fig. 6

OH lines in NGC 4418 (upper spectra) and Arp 220 (lower spectra). Λ-doubling is resolved in most transitions, except for the two narrowly separated 71 μm components that are blended into a single spectral feature. Model results are also shown. For NGC 4418, the blue, light-blue, and green curves show the models for the Ccore , Cwarm , and Cextended components, respectively; for Arp 220, the blue, light-blue, green, and yellow curves show the models for Cwest , Ceast , Cextended , and Chalo , respectively. The gray curves in panel f) show the expected contribution of the H2O 625 → 514 line to the spectra around 65.2 μm. The red curves show the total from all components.

Open with DEXTER
In the text
thumbnail Fig. 7

Spectra around the wavelengths of relevant O lines in NGC 4418 (upper profiles) and Arp 220 (lower profiles). Model predictions for NGC 4418 (Ccore) with , and for Arp 220 (Cwest) with are shown in red.

Open with DEXTER
In the text
thumbnail Fig. 8

Spectra around the wavelengths of relevant 18OH lines in NGC 4418 (upper profiles) and Arp 220 (lower profiles). Model predictions for NGC 4418 (Ccore) with , and for Arp 220 with are also shown. For the 18OH 120 μm doublet, the contributions by the different components in Arp 220 (see Fig. 6) are indicated; red denotes the total of all components.

Open with DEXTER
In the text
thumbnail Fig. 9

Spectra around the frequencies of HCN lines in NGC 4418 (upper profiles in each panel) and Arp 220 (lower profiles). Model predictions for NGC 4418 (Ccore) and Arp 220 (Cwest) are included.

Open with DEXTER
In the text
thumbnail Fig. 10

Spectra around the frequencies of relevant NH3 lines in NGC 4418 (upper profiles in each panel) and Arp 220 (lower profiles). Model predictions for NGC 4418 (Ccore) and Arp 220 (Cwest) are included.

Open with DEXTER
In the text
thumbnail Fig. 11

Line profiles of the [O i] 63 and 145 μm (a and b) transitions compared with those of some OH lines (c)f)). Upper histograms, NGC 4418; lower histograms, Arp 220. Model predictions are also included. For NGC 4418, the blue curves show the combined contribution of the Ccore and Cwarm components, and the green curves show the model for Cextended . For Arp 220, see caption of Fig. 6. Red denotes the total of all components.

Open with DEXTER
In the text
thumbnail Fig. 12

Comparison between the line shapes of the HCO+J = 3 → 2 (green histograms) and J = 4 → 3 (blue histograms) transitions toward a) the western and b) eastern nucleus of Arp 220 (from Sakamoto et al. 2009), with selected lines detected with Herschel/PACS (ci). The dashed vertical line indicates our adopted systemic velocity (Sect. 2.2). The velocity scale for the OH 79 (163) μm doublet is relative to the red (blue) Λ-component, and the systemic velocity for the other Λ-component is also indicated. The velocity of the O 303 → 212 line in c) is also indicated.

Open with DEXTER
In the text
thumbnail Fig. 13

Velocity shifts (ab), line widths (cd), and equivalent widths (ef) of the H2O lines in NGC 4418 (left) and Arp 220 (right). Panels gh show the line fluxes normalized to that of the 330 → 221 transition at 66.4 μm (see Sect. 2.3 for details). Error bars are 1-σ uncertainties from Gaussian fits to the lines.

Open with DEXTER
In the text
thumbnail Fig. 14

Velocity shifts, line widths, and equivalent widths of the [O i] and OH lines in NGC 4418 (black symbols) and Arp 220 (blue symbols). Error bars are 1-σ uncertainties from Gaussian fits to the lines.

Open with DEXTER
In the text
thumbnail Fig. 15

HCN line fluxes in NGC 4418 a) and Arp 220 b). In Arp 220, the fluxes of lines with Elower up to 570 K have been taken from SPIRE-FTS data (Rangwala et al. 2011). The curves and triangles show model predictions for the Ccore component of NGC 4418 and the Cwest component of Arp 220. The dotted curve in b) shows results obtained when only collisional excitation is included in the model. The solid curve in a) shows the best fit model obtained with τ200 ≈ 1, while the dashed and dotted curves correspond to τ200 = 4 with and without gas-dust mixing, respectively.

Open with DEXTER
In the text
thumbnail Fig. 16

Schematic representation of the modeled sources, showing the approximate spatial scales of the different far-IR components (see Fig. 1 and Tables 1 and 2). The Chot components in both galaxies (that account for the mid-IR spectra) are most probably associated with the nuclear regions and are not included here. The Chalo component in Arp 220 has no associated continuum, and is responsible for the absorption of the nuclear continuum (Ceast and Cwest) in the ground-state lines of H2O, OH, and O i. Although, due to its apparent low excitation, Chalo is represented here as the most extended component, its size is not constrained. The plot is an oversimplification of the actual models, where the different components are modeled separately to account approximately for non-spherical symmetry.

Open with DEXTER
In the text
thumbnail Fig. 17

Modeled absorbing flux in Arp 220 of several a) H2O, b) OH, c) O, d) NH3, and e) HCN lines as a function of the depth of the molecular shell measured from the surface of the continuum source, as parameterized by the dust opacity at 50 μm (τ50). Numbers in parenthesis indicate rounded wavelengths in μm. The total column of molecules is proportional to τ50, and we have adopted NH2O = 2 × 1018 × τ50 cm-2, and abundance ratios of OH/H2O = 0.5, , HCN/H2O = 0.3, and HCN/NH3 = 6. The dust temperature is 110 K.

Open with DEXTER
In the text
thumbnail Fig. 18

Modeled H2O line ratios in Arp 220 as a function of the dust temperature in the western nucleus. The dashed lines indicate the observed values, and the dotted lines the estimated upper and lower limits. In the lowest panel, the dotted line is the 3σ upper limit for the 818−707/330−221 flux ratio. Squares, triangles, circles, and crosses (only for Td,west = 90 K) show model results for “screen” H2O columns of NH2O = 7.5 × 1017, 1.5 × 1018, 3 × 1018, and 6 × 1018 cm-2, respectively.

Open with DEXTER
In the text
thumbnail Fig. 19

Predicted-to-observed H2O flux ratios in NGC 4418 (upper panel) and Arp 220 (lower panel) for the models including all components. See Sect. 3 for details.

Open with DEXTER
In the text
thumbnail Fig. 20

Modeled [O i] 63 μm emission above the continuum, integrated over blueshifted velocities (i.e. between −200 and 0 kms-1 with respect to the systemic velocity), for the nucleus of NGC 4418 as a function of the H2 density. The dashed line indicates the observed value.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.