Vol 17, No 8 (2017) / Jiao

Multi-transition study of the peculiar merger Arp 299

Multi-transition study of the peculiar merger Arp 299

Jiao Qian1, 2, 3, Zhu Ming1, 2,

National Astronomical Observatories, Chinese Academy of Sciences, Beijing 100012, China
Key Laboratory of Radio Astronomy, Chinese Academy of Sciences, Beijing 100012, China
University of Chinese Academy of Sciences, Beijing 100049, China

† Corresponding author. E-mail: mz@bao.ac.cn


Abstract: Abstract

We present a multi-transition study to investigate the physical properties of dust and molecular gas in the archetypical merger Arp 299 by using data including James Clerk Maxwell Telescope (JCMT) 850 and 450 μm observations, Herschel 500, 350, 250, 160 and 70 μm continuum maps, as well as the CO(3–2), CO(4–3) low-J CO lines and CO(11–10), CO(13–12), CO(14–13) high-J CO lines. The CO(3–2) and CO(4–3) lines are observed by JCMT, and the CO(11–10), CO(13–12), CO(14–13) lines are available on the Herschel Science Archive. The resolution of the Herschel Spectral and Photometric Imaging Receiver (SPIRE) Fourier transform spectrometer (FTS) CO(11–10) data is similar to that of the JCMT CO(3–2) line, while the resolution of the SPIRE/FTS CO(13–12) and Photodetector Array Camera and Spectrometer (PACS) CO(14–13) data is similar to that of JCMT CO(4–3), allowing us to obtain accurate line ratios of , and . By modeling the spectral energy distribution of the continuum data, we conclude that two components (cold and warm) exist in the dust, with the warm component occupying a small percent of the total dust mass. We further use a radiative transfer analysis code, RADEX, to calculate the density, temperature and column density of warm gas in the central region, which shows that the kinetic temperature is in the range 110 to 150 K and hydrogen density is in the range . We show that the hot dust is located in the central region of IC 694 with a radius of ∼ 4″ and estimate that the warm gas mass is in the range to , which contains 5.0%–15.0% of the total H2 mass for the region of IC 694. We also calculate the star formation rate of the galaxy in particular, which is much higher than that of the Milky Way.

Keywords: galaxies: merge;galaxies: Arp 299;galaxies: dust;galaxies: molecular gas;galaxies: warm center



1 Introduction

Luminous infrared galaxies (LIRGs), with , are galaxies having intense infrared emission (Soifer et al. 1987 ). Typically, LIRGs contain some of the most extreme star formation systems (Papadopoulos et al. 2012 ). Arp 299 (NGC 3690 + IC 694, VV 118, IRAS 11257+5850, Mrk 171,UGC6471/2) is one of the nearest (46 Mpc) LIRG ( (Sanders et al. 2003 )) mergers and has been the subject of detailed study. Arp 299 is made up of three main components and is dominated by intense star formation (Alonso-Herrero et al. 2000 ). Hα and near-infrared (NIR) images reveal extensive starburst activity in this system (García-Marín et al. 2006 ) and the star formation rate is about 70 (Sliwa et al. 2012 ). Single-dish observations of CO show that Arp 299 is very rich in molecular gas, and the 12CO/13CO line intensity ratios are extraordinarily high at both the J=1–0 and 2–1 transitions (e.g. Aalto et al. 1991 , Casoli et al. 1992 , Casoli et al. 1999 , Sliwa et al. 2012 ). Such a high ratio has also been found in a few LIRG mergers like Arp 220 and NGC 6240, but has never been observed in normal spiral galaxies (NSGs). This property indicates an extreme gas environment in these merging galaxies and may have a significant bearing on their starburst activity. Aperture synthesis HCN(1–0) maps of Arp 299 (Casoli et al. 1999 ) show that the line intensity ratio of HCN(1–0)/CO(1–0) is also much larger than that of NSGs at the nuclear region of IC 694, which signifies that IC 694 has a large amount of molecular gas in a dense phase because HCN (with a critical density ; Carolan et al. 2009 ) is a high density gas tracer. Mashian et al. ( 2015 ) showed that the low-J and high-J CO lines arise from two components with the same kinetic temperature of 200 K but different H2 number densities, typically and for IC 694, and the high-J CO lines are from the dense component. The physical conditions of molecular gas, especially high density gas traced by high-J CO lines, are crucial in understanding the star formation activity in this system. The main purpose of our study is to determine the physical properties of the warm and dense molecular gas in the galactic nucleus of IC 694, as well as the distribution and properties of dust in the complete system (Arp 299).

Throughout the paper, we use data from the James Clerk Maxwell Telescope (JCMT) (850, 450 μm, CO(3–2) and CO(4–3) lines), as well as data from the Herschel Spectral and Photometric Imaging Receiver (SPIRE; Griffin et al. 2010 ) and Photodetector Array Camera and Spectrometer (PACS; Poglitsch et al. 2010 ) on board the Herschel Space Observatory (Herschel; Pilbratt et al. 2010 ) (500, 350, 250, 160 and 70 μm, CO(11–10), CO(13–12), CO(14–13) lines). We study in detail the spectral energy distribution (SED) of the dust based on continuum data, as well as an excitation analysis of the warm center of IC 694 by using high-J CO data. In Section 2, we give a description of the observed data. The results and discussion are presented in Section 3. Section 4 is the conclusions.

2 Observations

2.1 The JCMT CO(3–2) Observations

We used the JCMT on Mauna Kea, Hawaii, to observe the CO(3–2) line emission toward the starburst galaxy Arp 299 during periods in July 1999 and June 2002. The receiver B3 was tuned to single side band (SSB) to observe the spectral line with the Digital Autocorrelation Spectrometer (DAS) at a 920 MHz bandwidth. There are 13 observed points with a sampling, which covers most of the dust emission region from Arp 299.

The 13 observed points of CO(3–2) were centered at right ascension (RA): and declination (Dec): 58°33′46″. For the center and near central points, the system temperatures are Tsys ∼ 350−450 K with for each point. While for the outer region, the system temperatures are Tsys ∼ 700 − 800 K and . During the grid observations, the center point at RA: and Dec: 58°33′46″ was repeatedly observed in order to verify pointing and line calibration. The reduced spectral profile with a calibration uncertainty of ∼ 10% is shown in Figure 1, where the pointing centers are marked as P1 to P13. Compared with data from the JCMT archive observed using the new multi-beam Heterodyne Array Receiver (HARP), we found that our data observed with the single-beam B3 receiver have higher signal-to-noise ratios (S/Ns) due to longer integration time at each point. Furthermore, the B3 data are more uniform, as all points were observed with the same receiver,whichmakes them ideal for quantitative analysis.

Fig. 1 The CO(3–2) data obtained for Arp 299 at a sampling of 7″ with the observed center in RA: , Dec: 58°33′46″. The observed points are overlaid on the CO(1–0) map (Casoli et al. 1999 ) with contours of 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0 and 8.0 Jy km s−1. P1 to P13 are centers of the observed points.

2.2 The JCMT CO(4–3) Observations

Observations of the CO(4–3) transition were conducted in May 2005, mainly on the nuclear part of IC 694. The receiver D3 was tuned to SSB to observe the CO(4–3) spectral line with the DAS at a 920 MHz bandwidth. There are a total of six observed points with a sampling for each point. They are centered at RA: , Dec: 58°33′47″. The system temperatures were Tsys ∼ 3000 − 4000 K with for each point. The reduced spectra are shown in Figure 2 with a spectral line calibration uncertainty of ∼ 15%, on which P1 to P6 are the respective centers of the observed points. The point P5 at RA: , Dec: 58°33′47″ has no obvious CO(4–3) emission line so we do not show its spectrum in the figure.

Fig. 2 The CO(4–3) data obtained for Arp 299 at a sampling of 5″ with the observed center in RA: , Dec: 58°33′47″. The observed positions are overlaid on the CO(1–0) map (Casoli et al. 1999 ) with contours of 1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0 and 8.0 Jy km s−1. P1 to P6 are the centers of the observed points.

2.3 Dust Continuum Submillimeter Observations with SCUBA

We used the Submillimeter Common-User Bolometric Array (SCUBA) to observe 450 and 850 μm of Arp 299 in 1998 March 28 and November 24 with grade 1 and 2 weather, respectively. The submillimeter bolometer array of SCUBA contains 91 bolometers at 450 μm and 37 at 850 μm, and covers a region ∼ 2.3′ in diameter. The secondary mirror was moved in a 64-step jiggle pattern with 1 second integration time for each point to provide fully sampled images. To remove background emissions which have a slowly varying gradient, the telescope nods with a chop throw of 60″ every 1 min in the standard jiggle pattern.

We used the SURF (Jenness & Lightfoot 1998 ) software package to reduce the data. Atmospheric absorption was corrected using opacities which were derived from sky dip measurements. Moreover, we flagged the noisy bolometers and removed large spikes. We mapped several calibration sources to flux calibrate the images in each night. The jiggle maps were co-added in the end, and we made a final map for the complete source Arp 299 after proper data reduction. For the 450 μm observations, we only used data which were taken in grade 1 weather with sufficient S/Ns. The final maps were in units of Jy beam−1 with the beam profiles acquired from the CRL618 beam maps. Typically, we estimate an uncertainty of 20% in the obtained fluxes at 850 and 450 μm.

2.4 Other Continuum and Spectroscopic Data

We also used the Herschel PACS 70, 160 μm and SPIRE 250, 350, 500 μm continuum maps, which are available on the Herschel Science Archive. The SPIRE data and PACS data were observed as part of the project “Star Formation And Activity In Infrared Bright Galaxies At 0 < z < 1.” We adopted a 9.5% (Bertincourt et al. 2016 ) calibration uncertainty for the extended SPIRE emission data. The calibration uncertainty for the PACS data was approximately 7% (e.g. Balog et al. 2014 ; Nielbock et al. 2013 ). The data were reduced by standard pipelines with Herschel Interactive Processing Environment (HIPE) software. Meanwhile, we also used the IRAM Plateau de Bure (PDB) interferometer CO(1–0) data (Casoli et al. 1999 ) and Herschel CO(11–10), CO(13–12), CO(14–13) data (Mashian et al. 2015 ), as well as the ISOCAM 15 μm (Gallais et al. 1999 ), NASA Infrared Telescope Facility (IRTF) 20, 25, 32 μm (Wynn-Williams et al. 1991 ) and Kuiper Widefield Infrared Camera (KWIC) 37 μm continuum data (Charmandaris et al. 2002 ). The details are shown in Charmandaris et al. ( 2002 , Table 1).

Position S 70 (Jy) S 160 (Jy) S 250 (Jy) S 350 (Jy) S 450 (Jy) S 500 (Jy) S 850 (Jy) T warm (K) T cold (K) M warm ( ) M cold ( )
A 90.22 42.57 3.27 0.31 89 ± 6 34 ± 1 6.0 ± 3.0 6.5 ± 0.8
B + C 47.35 26.27 2.44 0.24 33 ± 1 4.5 ± 0.6
Arp 299 136.20 66.59 19.49 6.09 5.66 2.16 0.60 97 ± 7 35 ± 1 8.1 ± 4.0 8.5 ± 0.6

Table 1 Fluxes, Dust Masses and Temperatures of Galaxies

3 Results

3.1 Dust Emission and Temperature

Figure 3 presents the Herschel and SCUBA continuum maps with the contours of CO(1–0) integrated intensity overlaid on them. Arp 299 is a merger system with two galaxies, e.g. IC 694 in the east and NGC 3690 in the west. Following the notation of Gehrz et al. ( 1983 ), we denote the nucleus of IC 694 as source A. NGC 3690 is composed of source B in the south and source C in the north. The top panels (SCUBA 450 μm on the left and SCUBA 850 μm on the right) and the middle panels (PACS 70 μm on the left and PACS 160 μm on the right) show that the dust emission of source A is slightly stronger than that of sources B and C. The PACS 70 and 160 μm emission in the nuclei looks quite similar to that of CO(1–0), especially the dust emission centered on the IC 694 nucleus. The SCUBA 450 and 850 μm emission in the nuclei also seems to be similar to the CO(1–0) peaks, with a small offset which may be caused by the pointing error or relatively poor resolution. It appears that most of the dust and molecular gas emissions come from the central region of Arp 299.

Fig. 3 The top panels are SCUBA maps at wavelengths 450 μm (left) and 850 μm (right). The middle panels are PACS 70 μm (left) and 160 μm (right), and the bottom panels are SPIRE 250 μm (left) and 350 μm (right). For comparison, all maps are overlaid with contours of CO(1–0) integrated intensity from the IRAM Plateau de Bure interferometer (Casoli et al. 1999 ). The contour levels are 1.0, 2.5, 4.0, 5.5, 7.0 and 8.5 Jy km s−1.

We fit the SED for each source as well for the Arp 299 system to derive the physical parameters (i.e. temperature and mass) of the dust. The resolutions of SCUBA 450 and 850 μm are 7.5″ × 7.5″ and 14.5″ × 14.5″, respectively. The PACS resolutions for 70 and 160 μm are 5.76″ × 5.46″ and 10.65″ × 12.13″, respectively. The SPIRE resolutions for 250, 350 and 500 μm are 18″ × 17″, 25″ × 23″ and 37″ × 34″, respectively. From Figure 3, we can see that in the SPIRE 250 μm data (the left panel at the bottom) the overlap region between source A and source B + C was not resolved well. Furthermore, the SPIRE 350 μm data and 500 μm data have even lower resolution, and it is hard to distinguish between IC 694 and NGC 3690 on these maps. Thus we only use the SCUBA 450, 850 μm data and the PACS 70, 160 μm data to fit the SED function for IC 694 and NGC 3690. Figure 3 also shows that source B has a region that is overlapped with source C and cannot be resolved very well, so we fit the region of source B+C.

Finally, we use a circle with a diameter of 44″, which is bigger than the resolution of all the observed data sets, to calculate the flux of Arp 299. From the maps in Figure 3, we can see that most of the dust emissions originate in this circle. The aperture correction factors for SCUBA 450 μm and SCUBA 850 μm are both 0.94 at diameter 44″ (Dempsey et al. 2013 ). The aperture correction factors are 0.87 and 0.82 for PACS 70 μm and PCAS 160 μm (Balog et al. 2014 ), and 1.26, 1.31 and 1.80 for SPIRE 250 μm, SPIRE 350 μm and SPIRE 500 μm at diameter 44″, respectively. When calculating the flux of source A and source B + C, we used circles with diameter 22″.

The aperture correction factors are both 0.75 for SCUBA 450 μm and SCUBA 850 μm at diameter 22″ (Dempsey et al. 2013 ). The aperture correction factors are 0.79 and 0.65 for PACS 70 μm and PCAS 160 μm (Balog et al. 2014 ), respectively. The measured fluxes of individual regions are presented in Table 1. Zhu et al. ( 2003 , 2007 ) have pointed out that the CO(3–2) line has significant contamination on the SCUBA 850 μm continuum fluxes. Such contamination can be estimated using the formula: This yields an average 25% − 30% flux correction for the Arp 299 system. Thus we apply 75% of the SCUBA 850 μm fluxes and use the corrected fluxes for the SED fitting.

Meanwhile, we also use the mid-infrared data of 15, 20, 25, 32 and 37 μm (Charmandaris et al. 2002 , table 1) to model the warm dust. The fluxes at 15, 20, 25, 32 and 37 μm for source A are 1.86, 5, 12.40, 27.0 ± 8.0 and 17.0 Jy, respectively, with 20% errors except for the 32 μm data. Note that mid-infrared data were not available for NGC 3690. The fluxes at 15, 20, 25, 32 and 37 μm for Arp 299 are 7.037, 11.90, 21.50, 43.00 and 36.96 Jy, respectively, with 20% errors. Assuming a Gaussian beam pattern for these mid-infrared data (Latvakoski et al. 1999 ), we can then ignore their aperture corrections. The mid-infrared fluxes mostly come from the warm and hot dust, which is usually distributed in the nuclear region of the galaxy. As shown in Section 3.2, we found that most of the warm dust component is located in the central 8″ region of IC 694. Thus here we assume most of the mid-infrared fluxes are distributed inside the circles used for the SED fitting. The fitted SED function is

where is the flux at frequency v, M is the dust mass, D = 46 Mpc is the distance, is the absorption cross section per unit dust mass from Draine ( 2003 ) with . and are the related Plank functions

Figure 4 presents the SED fitting results of source A (upper-left panel), source B + C (upper-right panel) and Arp 299 (bottom panel). For source A, the dust emissions have two components, a warm dust component with temperature 89 ± 6 K and mass , and a cold dust component with temperature 34 ± 1 K and mass . Because fewer data are available for source B + C, we fit the SED with a one-component model, which has a temperature 33 ± 1 K and dust mass . For Arp 299, a two-component model, consisting of a warm dust component with temperature 97 ± 7 K and dust mass , and a cold dust component with temperature 35 ± 1 K and mass , provides a good fit. The results are listed in Table 1.

Fig. 4 Upper-left panel: the SED fitting result of source A. Upper-right panel: the SED fitting result of source B + C. Bottom panel: the SED fitting result of Arp 299. The dotted lines represent the warm component, the dot-dashed lines represent the cold component and the solid lines are the best SED fitting results.

Rosenberg et al. ( 2014 ) used a dust model with three-components to fit the SED of Arp 299, and they gave a temperature 29.1 ± 3.5 K and mass for the cool component. For the warm component and hot component, they showed the temperatures are 60.6 ± 4.9 K and 239.4 ± 22.7 K, and masses are and 141 , respectively. Their hot dust component has very low mass and contributes very little to the mass budget of the whole system. Our results are consistent with their model with similar masses and temperatures for the cool components. Our results demonstrate that the majority of the dust mass comes from the cold dust component, but warm dust is necessary to explain the strong far-infrared (FIR) and submillimeter emission from this infrared luminous starbursting system.

3.2 The Warm Dust Component

In this section, we use the Herschel PACS 70 μm data to trace the distribution of the warm dust component on Arp 299. We further use the ratio to identify the dust temperature of the warm dust component. The Herschel data have been convolved to 7.5″ using the convolution kernels and programs described in Aniano et al. ( 2011 ) in order to match the resolution of the SCUBA 450 μm map. These are the highest resolution FIR and submillimeter data available for this system.

Figure 5 shows the contours of the ratio overlaid on the ratio map (upper-left panel), near-IR image at the K band (upper-right panel; Hibbard & Yun 1999 ) and PACS image at 70 μm (bottom panel). Higher ratios indicate higher dust temperatures. The NIR K band image clearly shows the location of the two nuclei IC 694 and NGC 3690. We can see that most of the warm dust emissions comes from the central region of IC 694 with a radius ∼ 4″, barely resolved at SCUBA 450 μm. Warm dust also exists in the galaxy NGC 3690, and the dust temperature on the southern disks appears to be slightly higher than that on the northern disks. However, the S/Ns of SCUBA 450 μm data are relatively low in these regions, thus we should be cautious when interpreting these data.

Fig. 5 The maps of , K band map and PACS 70 μm map with contours of . The contours are 13, 26, 40, 53 and 67. Upper-left panel is the color representation of with beam resolution of 7.5″. Upper-right panel is the K band map (Hibbard & Yun 1999, observed with the University of Hawaii 88 inch telescope) with beam resolution of 1″. Bottom panel is the PACS 70 μm map and its beam resolution is 5.8″.

The PACS 70 μm (Fig. 5 bottom panel) map has a resolution of 5.8″ and it clearly shows that most of the warm/hot dust is very centrally distributed around the nucleus of IC 694. This region also hosts the strongest CO(1–0) peak as shown in Figure 3. The strong mid-IR and FIR emissions indicate intense starburst activity at the nucleus. The physical conditions of molecular gas in this region are crucial in understanding the star formation and AGN activities of this merging system.

3.3 Physical Conditions of the Molecular Gas

As already shown in Figures 1 and 2, we have observed six points at CO(4–3) and 13 points at the CO(3–2) transition. We use the conversion factor to convert the flux unit from K km s−1 to Jy km s−1. The conversion factors are 25.2 and 50.4 for CO(3–2)and CO(4–3) with the equal to 0.62 and 0.31 for the JCMT CO(3–2) and CO(4–3) receivers, respectively. The pointing centers of the CO(3–2) observations are offset from the centers of sources A, B and C, hence we estimate the CO(3–2) fluxes of these sources by assuming that the CO(3–2) fluxes have a constant ratio with respect to the PACS 70 μm fluxes at each source. When calculating the ratio, we smoothed the resolution of the PACS 70 μm map to match the resolution of the CO(3–2) map. The detailed values are shown in Tables 2 and 3.

Source A P5 P1 P6 P10
CO(3–2) Flux (K km s−1) 36.30 26.26 53.47 48.92
PACS 70 μm (Jy) 63.48 44.43 78.10 48.60
CO(3–2)/Pacs70 0.57 0.59 0.68 1.01

Table 2 Comparison of the Observed Fluxes of Source A

Source B/C P2 P3 P4 P7 P8 P9 P11 P12 P13
CO(3–2) Flux (K km s−1) 24.71 16.41 17.5 44.35 29.83 32.27 25.33 21.14 28.27
PACS 70 μm (Jy) 15.13 11.06 11.23 29.40 21.27 24.28 14.33 15.85 21.85
CO(3–2)/Pacs70 0.89 1.00 1.12 0.95 1.08 1.07 0.94 1.10 0.84

Table 3 Comparison of the Observed Fluxes of Sources B and C

Table 2 shows that source A has almost constant ratios except for the point P10, which may be caused by pointing errors. We adopt a ratio of 0.63±0.08 using the mean value of points P5 and P6 (which are close to the center of source A), and then estimate the CO(3–2) flux of source A as 1233 Jy km s−1 within the CO(3–2) beam (14″).

Table 3 shows that for sources B and C, the flux ratios are roughly constant at about 1.07±0.04 (mean value of points P3, P4, P8 and P9) and 0.97 ± 0.19 (mean value of points P12 and P13), respectively. Thus the CO(3–2) fluxes of sources B and C are estimated as 833 Jy km s−1 and 627 Jy km s−1 within a diameter of 14″, respectively. The results are listed in Table 4.

Position RA Dec CO(4–3) Flux (Jy km s−1) CO(3–2) Flux (Jy km s−1)
A 58°33′47″ 2200 ± 220 1233 ± 180
B 58°33′41″ 833 ± 107
C 58°33′50″ 627 ± 139

Table 4 Center Fluxes for Galaxies in Arp 299

Sliwa et al. ( 2012 ) applied the radiative transfer code RADEX (van der Tak et al. 2007 ) to estimate the physical properties in the system using the CO(3–2), CO(2–1) and CO(1–0) transitions. They found that IC 694 and NGC 3690 consist of two gas components with a wide range of physical parameters, from a warm moderately dense gas with kinetic temperature and hydrogen density to a cold dense gas with and (Sliwa et al. 2012 ). Based on the CO(3–2)/(1–0) and CO(4–3)/(1–0) ratios, Papadopoulos et al. ( 2012 ) estimate the range of and using a one-phase large-velocity-gradient (LVG) model, and their best estimate is and . A moderately warm and dense gas is necessary to fit the CO(4–3) data. While the lowest J CO transitions can be fitted with a gas component with relatively low density and temperature, the high-J CO transitions can only be fitted with a gas density and a temperature range from medium to high (Rosenberg et al. 2014 ). Using the CO transition data from J = 1 to J = 14, Rosenberg et al. ( 2014 ) provided three photon-dominated region (PDR) models, in which the density of PDR I is in the range , that of PDR II is in the range and that of PDR III is in the range . They showed that the only way to reproduce the flux of the high-J CO lines is with a PDR model with a density and a radiation field , and the dust needs to be heated to a high temperature of 239.4 ± 22.7 K as well (Rosenberg et al. 2014 ). Mashian et al. ( 2015 ) used a two component LVG model to fit the IC 694 galaxy with the CO transition data from J = 1 to J = 20, and they found that the two components have the same kinetic temperature of 200 K but different H2 number densities of 103.4 and respectively.

Previous CO studies have indicated that there are at least two gas components in Arp 299, with different gas temperature and density ranges. Our own SED fitting of the dust emission also suggests at least two dust components as shown in Sections 3.1 and 3.2. The ratio map shows that the majority of the hot dust is from the central region of IC 694 with a diameter ∼ 8″. With our high resolution CO(4–3) and CO(3–2) data from JCMT, combined with the Herschel data at high-J CO transitions, we can have a better estimate of the physical parameters for the hot component of molecular gas traced by CO.

We assume that most of the high-J CO emissions are confined to a region similar to that of the hot dust, e.g. within the central 8″ in diameter, which can be treated as point sources with respect to the beam sizes of SPIRE/FTS and PACS. The resolution of the SPIRE/FTS CO(11–10) data is similar to that of the JCMT CO(3–2) line, while the resolution of the SPIRE/FTS CO(13–12) is similar to that of JCMT CO(4–3). Then the flux ratios of CO(11–10) to CO(3–2), and CO(13–12) to CO(4–3) can be calculated directly without beam convolution. The resolution of the Herschel PACS CO(14–13) data is also similar to that of the JCMT CO(4–3) line, and thus we can derive the line ratio as 0.07 ± 0.02 for source A, using the flux value of 1824 ± 273 Jy km s−1 for CO(14–13) (Mashian et al. 2015 ). However, it is likely that the CO(4–3) and CO(3–2) fluxes have contributions from both the cold and hot components, thus we need to correct for contamination of the cold component in the CO(4–3) and (3–2) emissions when using them to constrain the parameters for the hot components. We assume the cold gas component has a gas temperature similar to that of the dust temperature, e.g. about 30K. Then we can use the LVG solution of Papadopoulos et al. ( 2012 ) to estimate the parameters of the cold component as T = 30 K and density . Using the RADEX code, we then estimate that about 15% − 25% of the CO(4–3) and 35% − 45% of the CO(3–2) flux comes from the cold component. After removing the contribution from the cold component, the flux ratio of should be corrected as 0.14 ± 0.04, the ratio of should be 0.06 ± 0.01 and the flux ratio of should be 0.08 ± 0.02. These values are listed in Table 5, and we will use them to estimate the physical parameters for the hot component.

Position RA Dec CO(11–10) Flux (Jy km s−1) CO(13–12) Flux (Jy km s−1)
A 58°33′46″ 1418 ± 60 1126 ± 60 0.14 ± 0.04 0.06 ± 0.01
B 58°33′41″ 479 ± 40 405 ± 38 0.07 ± 0.02
C 58°33′48.2″ 488 ± 40 256 ± 40 0.10 ± 0.03

Table 5 Observed Fluxes and Flux Ratios of Arp 299

Our data are fitted with a 3D-grid RADEX model with a line width of 150 km s−1, kinetic temperature T kin range from 0 to 400 K, number density range from 102 to and column density N(CO) range from 1016 to . The results show that the flux ratio is not sensitive to the column density N(CO). The fitting contours of (upper-left panel), (upper-right panel) and (bottom panel) with column density are shown in Figure 6. Then the best fitting result is and according to the contours shown in Figure 6. Based on the RADEX RADEX_COLUMN program and the best fitting number density and kinetic temperature, the best fitting column density of molecular CO is in the range to .

Fig. 6 Contours calculated with the RADEX model with column density . Upper-left panel displays contours of / with kinetic temperature T kin and hydrogen density n(H2). Upper-right panel shows contours of / and bottom panel features contours of / .

The H2 column density is = according to equation , where is the abundance factor (Ward et al. 2003 ). In combination with the radius of the warm/hot center region, we can conclude that the center warm/hot H2 mass is in the range 3.8 × 107 to . Compared to the total H2 mass of for IC 694 (Sliwa et al. 2012 ), the warm/hot region contains 5.0%−15.0% of the total H2 mass for the region, indicating that most of the gas is cold and diffuse. Rigopoulou et al. ( 2002 ) found that the fraction of warm gas mass accounts for as much as 10% of the total galactic mass in starbursts and 2%−35% in Seyferts, which is consistent with our result.

3.4 Kinematic Analysis

The gas in the central region of IC 694 has both a high temperature T kin (range 110−150 K) and high density (range ). Such conditions imply a high gas pressure. The outer region has a relatively low density = 104 cm−3 and kinetic temperature K (Papadopoulos et al. 2012 ). To maintain hydrostatic equilibrium, we have

where represents the enclosed mass at radius r, p is the pressure, dr is the thickness of the transition layer, G is the gravitational constant and ρ(r) is the density of the gas at radius r. We can deduce that
where n is the number density, k is the Boltzmann constant, T warm/hot is the temperature of the warm/hot gas and T cold is the temperature of the cold gas. The warm/hot temperature is 110 K, the density n is and the number density is for ρ(r). Assuming the thickness of the transition layer dr is about 1/10r, then the mass that is required to maintain hydrostatic equilibrium is at the 4″ radius of the warm/hot gas region (assuming a distance of 46 Mpc). If the thickness is smaller, then the required mass of the warm/hot gas will be larger. This value is similar to the total mass of the warm/hot gas component we estimated with the RADEX model ( ). Considering other mass components such as stellar mass, there appears to be enough mass to provide the gravity necessary to confine the central hot and dense gas against thermal pressure to maintain hydrostatic equilibrium. Thus the hot gas component could be stable and could last for a long time at the nucleus of IC 694.

3.5 Infrared Luminosity and Star Formation Rate

The infrared luminosities are derived from the SED fitting results, and are shown in Table 6. The star formation rates (SFRs) can be deduced from infrared luminosities by using the equation below (Kennicutt 1998 )

with . The resulting SFR is 86.5 for source A and 36.3 for source B + C based on Equation (5). The SFR of Arp 299 is 150.5 , much higher than the result of Sliwa (70 ), which may be associated with the different methods of calculating infrared luminosities. In previous studies, L IR was calculated by using IRAS S 60 μm and S 100 μm data due to the lack of submillimeter data. Thanks to the Herschel and SCUBA data, we can model the SED more accurately now, which leads to more accurate integrated infrared luminosities and SFRs. For source B + C, the one component model may underestimate the infrared luminosity and the SFR. For comparison, the SFR is 5 in the Milky Way (Smith et al. 1978 ), so the SFR in Arp 299 is significantly higher than that in normal galaxies.

Position L IR ( ) SFR ( )
A 5.0 ± 3.2 86.5 ± 55.3
B + C 2.1 ± 1.6 36.3 ± 27.7
Arp 299 8.7 ± 5.7 150.5 ± 98.6

Table 6 Luminosity and SFR for Each Galaxy

4 Summary and Conclusions

We present the observed JCMT CO(3–2) and CO(4–3), and SCUBA 850 and 450 μm data, as well as the fully-sampled Herschel 500, 350, 250, 160 and 70 μm continuum maps of Arp 299. Combined with the mid-infrared data at 5, 20, 25, 32 and 37 μm (Charmandaris et al. 2002 ), we fit the SED of the dust emission and find that there are at least two dust components (a cold component and a warm component) in source A and Arp 299. We find that most of the dust is cold, and warm dust accounts for a small amount of the total dust mass. The high resolution ratio map shows that the warm/hot dust is mostly distributed in the central region of source A with radius ∼ 4″.

By combining the Herschel CO(11–10), CO(13–12) and CO(14–13) data with the JCMT CO(3–2) and CO(4–3) data, which have similar spatial resolution and beam size, we have used a RADEX model to calculate the parameters for the warm/hot molecular gas component in the center of IC 694. We estimate that T kin is in the range of 110 to 150 K and . The mass of the warm/hot gas M (H2) is , which accounts for 5.0%−15.0% of the total H2 mass of IC 694. The star formation rate is 86.5 for source A, 36.3 for source B + C, and 150.5 for Arp 299.


References

Aalto S. Johansson L. E. B. Booth R. S. Black J. H. 1991 A&A 249 323
Alonso-Herrero A. Rieke G. H. Rieke M. J. Scoville N. Z. 2000 ApJ 532 845
Aniano G. Draine B. T. Gordon K. D. Sandstrom K. 2011 PASP 123 1218
Balog Z. Müller T. Nielbock M. et al. 2014 Experimental Astronomy 37 129
Bertincourt B. Lagache G. Martin P. G. et al. 2016 A&A 588 A107
Carolan P. B. Khanzadyan T. Redman M. P. et al. 2009 MNRAS 400 78
Casoli F. Dupraz C. Combes F. 1992 A&A 264 55
Casoli F. Willaime M.-C. Viallefond F. Gerin M. 1999 A&A 346 663
Charmandaris V. Stacey G. J. Gull G. 2002 ApJ 571 282
Dempsey J. T. Friberg P. Jenness T. et al. 2013 MNRAS 430 2534
Draine B. T. 2003 ARA&A 41 241
Gallais P. Laurent O. Charmandaris V. et al. 1999 ESA Special Publication 427 The Universe as Seen by ISO Cox P. Kessler M. 881
García-Marín M. Colina L. Arribas S. Alonso-Herrero A. Mediavilla E. 2006 ApJ 650 850
Gehrz R. D. Sramek R. A. Weedman D. W. 1983 ApJ 267 551
Griffin M. J. Abergel A. Abreu A. et al. 2010 A&A 518 L3
Hibbard J. E. Yun M. S. 1999 AJ 118 162
Jenness T. Lightfoot J. F. 1998 Astronomical Society of the Pacific Conference Series 145 Astronomical Data Analysis Software and Systems VII Albrecht R. Hook R. N. Bushouse H. A. 216
Kennicutt R. C. Jr. 1998 ARA&A 36 189
Latvakoski H. M. Stacey G. J. Gull G. E. Hayward T. L. 1999 ApJ 511 761
Mashian N. Sturm E. Sternberg A. et al. 2015 ApJ 802 81
Nielbock M. Müller T. Klaas U. et al. 2013 Experimental Astronomy 36 631
Papadopoulos P. P. van der Werf P. P. Xilouris E. M. et al. 2012 MNRAS 426 2601
Pilbratt G. L. Riedinger J. R. Passvogel T. et al. 2010 A&A 518 L1
Poglitsch A. Waelkens C. Geis N. et al. 2010 A&A 518 L2
Rigopoulou D. Kunze D. Lutz D. Genzel R. Moorwood A. F. M. 2002 A&A 389 374
Rosenberg M. J. F. Meijerink R. Israel F. P. et al. 2014 A&A 568 A90
Sanders D. B. Mazzarella J. M. Kim D.-C. Surace J. A. Soifer B. T. 2003 AJ 126 1607
Sliwa K. Wilson C. D. Petitpas G. R. et al. 2012 ApJ 753 46
Smith L. F. Biermann P. Mezger P. G. 1978 A&A 66 65
Soifer B. T. Sanders D. B. Madore B. F. et al. 1987 ApJ 320 238
van der Tak F. F. S. Black J. H. Schöier F. L. Jansen D. J. van Dishoeck E. F. 2007 A&A 468 627
Ward J. S. Zmuidzinas J. Harris A. I. Isaak K. G. 2003 ApJ 587 171
Wynn-Williams C. G. Hodapp K.-W. Joseph R. D. et al. 1991 ApJ 377 426
Zhu M. Gao Y. Seaquist E. R. Dunne L. 2007 AJ 134 118
Zhu M. Seaquist E. R. Kuno N. 2003 ApJ 588 243
Cite this article: Jiao Qian, Zhu Ming. Multi-transition study of the peculiar merger Arp 299. Cancer Biol Med. 2017; 8:077.

Refbacks

  • There are currently no refbacks.