Comparison of bar strengths in active and non-active galaxies

Eija Laurikainen 0 P Heikki Salo 0 Pertti Rautiainen 0 0 Division of Astronomy, Department of Physical Sciences , FIN-90014, Finland From an original sample of 107 spiral galaxies, the bar strengths of 21 active galaxies are compared with those of 22 non-active galaxies. Our identifications of bars are inferred from near-IR images using Fourier methods. The bar torques are determined using a new technique due to Buta & Block, in which tangential forces are calculated in the bar region normalized to the axisymmetric radial force field. As a data base we use the JHK images of the 2 Micron All Sky Survey. The ellipticities e of the bars are also estimated with an isophotal fitting algorithm and the bar lengths from the phases of m 2 and m 4 Fourier components of density. We show a first clear indication that the ellipticity of a bar, generally used as a measure of the bar strength, is quite well correlated with the maximum relative tangential force, Qb, in the bar region. Most surprisingly, the galaxies with the strongest bars are non-active. A possible understanding of this unusual result is that previous gaseous inflow in such cases may have been so efficient that fuelling of the active nuclei may simply have ceased. We find that nuclear activity occurs preferentially in those barred early-type galaxies in which the maximal bar torques are weak kQbl 0:21 and appear at quite large distances from the galactic centre when scaled with the radial scalelength of the disc krQb /hl 1:24 . For comparison, for the non-active galaxies kQbl 0:37 and krQb /hl 0:59. The force maximum in the active late-type galaxies also appears at rather large distances, but the difference from the non-active galaxies is smaller. These results imply that the bulges may be important for the onset of nuclear activity, but it is not clear why nuclear activity appears in some early-type galaxies but is missing from some others. We also find that for active early-type galaxies, bar length is not correlated with bar strength, although a weak correlation appears for the other barred galaxies studied. I N T R O D U C T I O N It is now widely accepted that non-axisymmetric forces are needed to trigger nuclear activity by accreting gas to the central regions of galaxies. Primary bars are known to be efficient for dragging gas on scales larger than 1 kpc, but some other mechanism is needed to allow this gas finally to fall into the active nucleus, provided for example by secondary bars (Shlosman, Frank & Begelman 1989). Indeed, star formation is found to be enhanced in barred galaxies (Martinet & Friedli 1997; Aquerri 1999), but the connection between bars and nuclear activity (AGNs in terms of accretion discs and black holes) is not clear observationally. For example, there is no indisputable agreement showing that the number of bars is larger for AGNs in comparison with non-active systems. An excess of bars in AGNs is found by some authors (Knapen, Shlosman & Peletier 2000; Laine et al. 2001), while most studies give similar bar fractions for active and non-active galaxies (McLeod & Rieke 1995; Moles et al. 1995; Ho, Filippenko & Sargent 1997; Mulchaey & Regan 1997; Hunt & Malkan 1999; Marquez et al. 2000). Bars can act as driving forces for the central activity via the strong inflow of gas in shock regions associated with the rotating bar potential. Not surprisingly, the nuclear regions of barred galaxies have on average higher concentrations of molecular gas than normal galaxies (Sakamoto, Baker & Scoville 2000; Sheth 2001), which makes understandable the connection found between bars and high star-formation activity. Star-formation activity is found to be correlated with the properties of bars, being enhanced especially in long bars with high ellipticities, usually regarded as strong bars (Martinet & Friedli 1997; Aquerri 1999). However, not all long bars have pronounced current star-formation activity. On the other hand, it has been suggested that Seyferts may even avoid strong bars (Shlosman, Peletier & Knapen 2000; Laine et al. 2001), but this has been questioned by Marquez et al. (2000), who argued that both the lengths and strengths of the primary bars are similar for Seyferts and for non-Seyfert galaxies. In all these works bar strength is estimated indirectly from the ellipticity of a bar, based on analytical work by Athanassoula (1992a). The size of a bar is related to the Hubble type so that bars in early-type systems are generally longer than in late-type systems, when normalized to the galaxy diameter at 25 magnitude isophote D25 (Elmegreen & Elmegreen 1985; Duval & Monnet 1985; Martin 1995; Regan & Elmegreen 1997). Also, Elmegreen & Elmegreen (1985) find a weak correlation between the bar axial ratio and the Hubble type, but this has not been confirmed by Martin (1995) using a larger sample of galaxies. Elmegreen & Elmegreen (1985) also showed that bars in early-type galaxies are generally flat, while those in late-type systems are exponential. Altogether, the Hubble type is expected to be an important factor in controlling the properties of bars and probably also the inflow of gas. The ellipticity of a bar is not a full description of its strength, depending also on the mass of the bar. Moreover, the relative perturbation associated with the bar depends on the central force field, i.e. the presence of a massive bulge. Therefore bar strengths are better evaluated by the tangential forces normalized to the total axisymmetric force fields, as suggested by Combes & Sanders (1981) and applied for galaxies by Buta & Block (2001). We use a similar approach and determine bar torques for 107 spiral galaxies in JHK-bands. Fourier analysis is used to identify bars, and for those galaxies with well-defined bars, strengths are compared between the active and non-active galaxies. We also test how well the bar strength and the ellipticity of a bar are correlated. T H E S A M P L E A N D T H E M E T H O D The original sample consists of those spiral galaxies with BT , 12:5 mag, cz , 2500 km sec21 and i , 670 in the Third Reference Catalog of Bright Galaxies (de Vaucouleurs et al. 1991, hereafter RC3), for which high-quality images were available in the 2 Micron All Sky Survey (hereafter 2MASS). By i we denote the inclination of the galactic disc. Additionally, some of the weakest objects were eliminated so that the number of galaxies in the sample was 107. In the active galaxy category we include Seyferts, LINERs and H II galaxies, for which the spectral classifications were taken from the NASA/IPAC Extragalactic Database (NED), where the latest spectral classifications are available. Altogether, 53 of the galaxies show nuclear activity and 31 are classified as barred (SB) in RC3. The image resolution is one arcsecond per pixel and the H-images are generally deeper than the images in the J- or K-bands. The selection effects of the sample are discussed by Laurikainen & Salo (2002). We found that the (...truncated)