We present a detailed analysis of deep far-infrared observations of the nearby edge-on star-forming galaxy NGC 4631 obtained with the Herschel Space Observatory. Our PACS images at 70 and 160 um show a rich complex of filaments and chimney-like featu
res that extends up to a projected distance of 6 kpc above the plane of the galaxy. The PACS features often match extraplanar Halpha, radio-continuum, and soft X-ray features observed in this galaxy, pointing to a tight disk-halo connection regulated by star formation. On the other hand, the morphology of the colder dust component detected on larger scale in the SPIRE 250, 350, and 500 um data matches the extraplanar H~I streams previously reported in NGC 4631 and suggests a tidal origin. The PACS 70/160 ratios are elevated in the central ~3.0 kpc region above the nucleus of this galaxy (the superbubble). A pixel-by-pixel analysis shows that dust in this region has a higher temperature and/or an emissivity with a steeper spectral index (beta > 2) than the dust in the disk, possibly the result of the harsher environment in the superbubble. Star formation in the disk seems energetically insufficient to lift the material out of the disk, unless it was more active in the past or the dust-to-gas ratio in the superbubble region is higher than the Galactic value. Some of the dust in the halo may also have been tidally stripped from nearby companions or lifted from the disk by galaxy interactions.
Powerful winds driven by active galactic nuclei (AGN) are often invoked to play a fundamental role in the evolution of both supermassive black holes (SMBHs) and their host galaxies, quenching star formation and explaining the tight SMBH-galaxy relati
ons. Recent observations of large-scale molecular outflows in ultra-luminous infrared galaxies (ULIRGs) have provided the evidence to support these studies, as they directly trace the gas out of which stars form. Theoretical models suggest an origin of these outflows as energy-conserving flows driven by fast AGN accretion disk winds. Previous claims of a connection between large-scale molecular outflows and AGN activity in ULIRGs were incomplete because they were lacking the detection of the putative inner wind. Conversely, studies of powerful AGN accretion disk winds to date have focused only on X-ray observations of local Seyferts and a few higher redshift quasars. Here we show the clear detection of a powerful AGN accretion disk wind with a mildly relativistic velocity of 0.25c in the X-ray spectrum of IRAS F11119+3257, a nearby (z = 0.189) optically classified type 1 ULIRG hosting a powerful molecular outflow. The AGN is responsible for ~80% of the emission, with a quasar-like luminosity of L_AGN = 1.5x10^46 erg/s. The energetics of these winds are consistent with the energy-conserving mechanism, which is the basis of the quasar mode feedback in AGN lacking powerful radio jets.
We have carried out extensive and detailed photoionization modeling to successfully constrain the locations of different emission-line galaxies in optical and mid-infrared diagnostic diagrams. Our model grids cover a wide range in parameter space for
the active galaxy continuum and starburst galaxies with different stellar population laws and metallicities. We compare the predicted AGN and star-formation mid-infrared line ratios [Ne III]15.56mm/[Ne II]12.81mm and [O IV]25.89mm/[Ne III]15.56mm to the observed values, and find that the best fit for the AGN is via a two-zone approximation. This two-zone approximation is a combination of a matter-bounded component, where [Ne III] and [O IV] are emitted efficiently, and a radiation-bounded component that maximizes [Ne II] emission. We overlay the predictions from this two-zone approximation onto the optical [O III]l5007/Hbeta and [N II]l6583/Halpha diagnostic diagram derived from the Sloan Digital Sky Survey, to find that the high-density and low-ionization radiation-bounded component in our two-zone AGN approximation model provides a good lower limit for [N II] emission. This establishes a new theoretical demarcation line for the minimum AGN contribution in this diagram. This new classification results by a factor of ~1.4 in a higher AGN population than predictions derived from previous divisions of star-forming galaxies. Similarly, we define a maximum AGN contribution in the [O III]/Hbeta and [N II]/Halpha diagram by using a two-zone approximation within a parameter range typical of the narrow-line region.
We compare mid-infrared emission-line properties, from high-resolution Spitzer spectra of a hard X-ray (14 -- 195 keV) selected sample of nearby (z < 0.05) AGN detected by the Burst Alert Telescope (BAT) aboard Swift. The luminosity distribution for
the mid-infrared emission-lines, [O IV] 25.89 micron, [Ne II] 12.81 micron, [Ne III] 15.56 micron and [Ne V] 14.32/24.32 micron, and hard X-ray continuum show no differences between Seyfert 1 and Seyfert 2 populations, however six newly discovered BAT AGNs are under-luminous in [O IV], most likely the result of dust extinction in the host galaxy. The overall tightness of the mid-infrared correlations and BAT fluxes and luminosities suggests that the emission lines primarily arise in gas ionized by the AGN. We also compare the mid-infrared emission-lines in the BAT AGNs with those from published studies of ULIRGs, PG QSOs, star-forming galaxies and LINERs. We find that the BAT AGN sample fall into a distinctive region when comparing the [Ne III]/[Ne II] and the [O IV]/[Ne III] ratios. These line ratios are lower in sources that have been previously classified in the mid-infrared/optical as AGN than those found for the BAT AGN, suggesting that, in our X-ray selected sample, the AGN represents the main contribution to the observed line emission. These ratios represent a new emission line diagnostic for distinguishing between AGN and star forming galaxies.
We have studied the relationship between the high- and low-ionization [O IV] lambda 25.89 micron, [Ne III] lambda 15.56 micron and [Ne II] lambda 12.81 micron emission lines with the aim of constraining the active galactic nuclei (AGN) and star forma
tion contributions for a sample of 103 Seyfert galaxies. We used the [O IV] and [Ne II] emission as tracers for the AGN power and star formation to investigate the ionization state of the emission-line gas. We find that Seyfert 2 galaxies have, on average, lower [O IV]/[Ne II] ratios than those of Seyfert 1 galaxies. This result suggests two possible scenarios: 1) Seyfert 2 galaxies have intrinsically weaker AGN, or 2) Seyfert 2 galaxies have relatively higher star formation rates than Seyfert 1 galaxies. We estimate the fraction of [Ne II] directly associated with the AGN and find that Seyfert 2 galaxies have a larger contribution from star formation, by a factor of ~1.5 on average, than what is found in Seyfert 1 galaxies. Using the stellar component of [Ne II] as a tracer of the current star formation we found similar star formation rates in Seyfert 1 and Seyfert 2 galaxies. We examined the mid- and far-infrared continua and find that [Ne II] is well correlated with the continuum luminosity at 60 micron and that both [Ne III] and [O IV] are better correlated with the 25 micron luminosities than with the continuum at longer wavelengths, suggesting that the mid-infrared continuum luminosity is dominated by the AGN, while the far-infrared luminosity is dominated by star formation. Overall, these results test the unified model of AGN, and suggest that the differences between Seyfert galaxies cannot be solely due to viewing angle dependence.
We have studied the relationship between the [O IV] lambda 25.89 micron emission line luminosities, obtained from Spitzer spectra, the X-ray continua in the 2-10 keV band, primarily from ASCA, and the 14-195 keV band obtained with the SWIFT/Burst Ale
rt Telescope (BAT), for a sample of nearby (z < 0.08) Seyfert galaxies. For comparison, we have examined the relationship between the [O III] 5007, the 2-10 keV and the 14-195 keV luminosities for the same set of objects. We find that both the [O IV] and [O III] luminosities are well-correlated with the BAT luminosities. On the other hand, the [O III] luminosities are better-correlated with 2-10 keV luminosities than are those of [O IV]. When comparing [O IV] and [O III] luminosities for the different types of galaxies, we find that the Seyfert 2s have significantly lower [O III] to [O IV] ratios than the Seyfert 1s. We suggest that this is due to more reddening of the narrow line region (NLR) of the Seyfert 2s. Assuming Galactic dust to gas ratios, the average amount of extra reddening corresponds to a hydrogen column density of ~ few times 10^21 cm^-2, which is a small fraction of the X-ray absorbing columns in the Seyfert 2s. The combined effects of reddening and the X-ray absorption are the probable reason why the [O III] versus 2-10 keV correlation is better than the [O IV] versus 2-10 keV, since the [O IV] emission line is much less affected by extinction. Overall, we find the [O IV] to be an accurate and truly isotropic indicator of the power of the AGN. This suggests that it can be useful in deconvolving the contribution of the AGN and starburst to the spectrum of Compton-thick and/or X-ray weak sources.
