Infrared identification of IGR J09026-4812 as a Seyfert 1 galaxy
AAstronomy & Astrophysics manuscript no. igrj09026˙arx c (cid:13)
ESO 2018November 8, 2018
Infrared identification of IGR J09026 − (cid:63) J.A. Zurita Heras , S. Chaty , and J. A. Tomsick Laboratoire AIM, CEA / DSM-CNRS-Universit´e Paris Diderot, IRFU / Service d’Astrophysique, 91191 Gif-sur-Yvette, Francee-mail: [email protected] ; e-mail: [email protected] Space Sciences Laboratory, 7 Gauss Way, University of California, Berkeley, CA 94720 − [email protected] Received –; accepted –
ABSTRACT
Context.
IGR J09026 − INTEGRAL in 2006 as a new hard X-ray source. Thereafter, an observation with
Chandra pinpointed a single X-ray source within the ISGRI error circle, showing a hard spectrum, and improving its high-energy lo-calisation to a subarcsecond accuracy. Thus, the X-ray source was associated to the infrared counterpart 2MASS J09023731 − JHK S photometry indicated a highly reddened source. The high-energy properties and the counterpart photometry suggesteda high-mass X-ray binary with a main sequence companion star located 6.3–8.1 kpc away and with a 0.3–10 keV luminosity of8 + − × ergs s − . Aims.
New optical and infrared observations were needed to confirm the counterpart and to reveal the nature of IGR J09026 − Methods.
We performed optical and near infrared observations on the counterpart 2MASS J09023731 − / NTTtelescope on March 2007. We achieved photometry and spectroscopy in near infrared wavelengths and photometry in optical wave-lengths.
Results.
The accurate astrometry at both optical and near infrared wavelengths confirmed 2MASS J09023731 − − λ . µ m line, and theHI Pa β and Pa α lines, typical in galaxies with an active galactic nucleus. The broadness of these lines reached values as large as4000 km s − pointing towards a type 1 Seyfert galaxy. The redshift of the source is z = . ± . − Key words.
X-rays: individual: IGR J09026 −
1. Introduction
The ESA high-energy space mission
INTEGRAL (Winkler et al.2003) has performed a deep survey of our Galaxy allowing thediscovery of many new hard X-ray sources. This has been mainlyachieved with the coded-mask imager IBIS / ISGRI (15 keV–1 MeV, field of view of 29 ◦ × ◦ , angular resolution of 12 (cid:48) ;Ubertini et al. 2003; Lebrun et al. 2003). The 20–100 keV hardX-ray emission as observed by ISGRI has been regularly re-ported (Bird et al. 2006, 2007; Bodaghee et al. 2007). The un-classified source IGR J09026 − = ◦ .
638 and Dec. = − ◦ .
196 (1 . (cid:48) at 68% confidence level). The average 20–40and 40–100 keV fluxes were 0 . ± . . ± . ff ective exposure of 1.35 Ms. These val-ues only su ff ered a slight change in the 3rd ISGRI catalogue(Bird et al. 2007): R.A. (2000) = ◦ . = − ◦ . . (cid:48) at 90% confidence level), F − = . ± . F − = . ± . ff ective expo-sure of 1.5 Ms. The source IGR J09026 − ROSAT counterpart located 2 . (cid:48) awayfrom the ISGRI position at R.A. (2000) = h m . s andDec. = − ◦ (cid:48) . (cid:48)(cid:48) (20 (cid:48)(cid:48) ) that showed an average count rateof (2 . ± . × − counts s − (Stephen et al. 2006). (cid:63) Based on observations made with ESO Telescopes at the La SillaObservatory under programme ID 078.D − Tomsick et al. (2008) reported a 5 ks
Chandra observa-tion performed on Feb. 5, 2007, that showed a single X-raysource within the ISGRI error circle located at R.A. (2000) = h m . s and Dec. = − ◦ (cid:48) . (cid:48)(cid:48) (0 . (cid:48)(cid:48) at 90% confidencelevel). This position is located outside the ROSAT error circle,discarding the soft X-ray counterpart suggested by Stephen et al.(2006). The 0.3–10 keV spectrum was fitted with an absorbedpower law model: N H = (1 . + . − . ) × cm − , Γ = . + . − . (90% confidence errors) and an unabsorbed 0.3–10 keV flux of1 . × − erg cm − s − . They also showed that the source wasnot intrinsically absorbed comparing the fitted N H to the atomicand molecular hydrogen column densities through our Galaxy.They suggested the source 2MASS J09023731 − . (cid:48)(cid:48) away from the Chandra position. Its IR magnitudes were J = . ± . H = . ± .
07, and K S = . ± .
04. Due toits K S brightness, high extinction, and hard power law, Tomsicket al. (2008) suggested that IGR J09026 − − a r X i v : . [ a s t r o - ph . C O ] M a y J.A. Zurita Heras et al.: Infrared identification of IGR J09026 − J E N chandra rosat H E N chandra rosat Ks E N chandra rosat
Fig. 1.
Left to right, top to bottom: VRI and
JHK s images taken on March 10 (NIR) and 14 (optical), 2007.
2. Observations and data analysis
The observations were carried out with the ESO 3.6 m NewTechnology Telescope (NTT) at La Silla Observatory, Chile,as part of the program 078.D–0268(B) through service mode.The optical and NIR observations were achieved with the im-ager SUSI–2 (350–900 nm) and the spectro-imager SOFI (0.9–2.4 µ m), respectively. Both instruments are installed on the sameNasmyth focus of the NTT. Astrometry, photometry and spec-troscopy were achieved during these observations. NIR photometry in the bands J , H and K S were performed onMarch 5 and 10, 2007. The observations were centred on the ac-curate X-ray position of IGR J09026 − . (cid:48)(cid:48) / pixel and a field of view of 4 . (cid:48) × . (cid:48) .Nine images were taken for each filter J , H and K S with integra-tion times of 60 s each. For each filter, four of the nine imageswere taken with a slight o ff set of ∼ (cid:48)(cid:48) that allowed us to buildthe NIR sky in order to subtract it from the images. Two standardstars chosen in the faint NIR standard star catalogue (Perssonet al. 1998) were also observed: S255 − S and S262 − E. Five ob-servations per filter were performed on each standard star. Thefirst observation was centred on the target, and then, the next fourimages were taken with an o ff set of ∼ (cid:48)(cid:48) compared to the firstone. The same strategy with the same standard stars was appliedduring the 2nd night. The seeing conditions varied from ∼ . (cid:48)(cid:48) during the 1st night to ∼ . (cid:48)(cid:48) during the 2nd night.NIR spectroscopy using both low-resolution Blue and Redgrisms was carried out on March 5 and 12, 2007. For each filter,twelve spectra were taken with the 1 . (cid:48)(cid:48) slit. Half of the spectrawere taken on the source and the other half with an o ff set of60 (cid:48)(cid:48) in order to subtract the NIR sky from the images, the jitterpattern being AAA BBB BBB AAA. The integration time was180 s for each spectrum for a total observation of 4320 s. Twospectroscopic standard stars were observed: HD62388 (spectraltype (sp.T.) A0V) and Hip0842847 (sp.T. G2V) on March 5 and 12, respectively. For each standard star, four spectra in both Blueand Red grisms were taken, of which half of them were takenwith an o ff set of 45 (cid:48)(cid:48) . Only optical photometry was carried out on March 14, 2007,with the filters U , B , V , R , I , Z . One image per filter was obtainedwith a field of view of 5 . (cid:48) × . (cid:48) with a binning factor of 2 thatimplied a pixel scale of 0.161 (cid:48)(cid:48) / pixel. The integration time was60 s in each filter. Thirteen photometric standard stars selectedin the optical standard star catalogue of Landolt (1992) were ob-served in the 2 fields 98 (733,1087,1102,1112,1119,1122,1124)and RU 152 + RU 152(A,B,C,E,F). The integration times variedbetween 5 and 30 s.
The reduction of both optical and NIR data was performedwith the Image Reduction and Analysis Facility (IRAF ) version2.13beta2. Data reduction was performed using standard proce-dures on the optical and NIR images, including the crosstalk,the correction of the dark current (in optical), the flat-fieldingand the NIR sky subtraction.We performed accurate astrometry on each image( U , B , V , R , I , Z , J , H , K S ) using the gaia tool from the Starlinksuite and using the 2MASS catalogue for the NIR images andthe USNO B1.0 catalogue for the optical images. The NIRroot mean square (rms) of the astrometrical fit was alwayslower than 0 . (cid:48)(cid:48) with the expected pixel scale in x,y axis of − . × . (cid:48)(cid:48) / pixel. The optical rms of the astrometrical fitwas also lower than 0 . (cid:48)(cid:48) with the expected pixel scale in x,yaxis of 0 . × . (cid:48)(cid:48) / pixel.We carried out aperture photometry in a crowded fieldusing the IRAF digiphot.daophot package. For the NIRphotometry, the point sources located near IGR J09026 − IRAF is available at http://iraf.net/ .A. Zurita Heras et al.: Infrared identification of IGR J09026 − try on the source was performed with several ellipses usingthe stsdas.isophot and digiphot.apphot packages. Theinstrumental magnitudes m instr were transformed into apparentmagnitudes m app using the standard photometric relation: m app = m instr − Z p − ext × AM , where Z p is the zero-point, ext the ex-tinction and AM the airmass. The colour term was not used inthe NIR because there were not enough standard stars nor at op-tical wavelengths because we did not detect the sources of inter-est in several bands, particularly in V . The Z p parameters werefitted using the photometric relation in order to match the in-strumental and apparent magnitudes of the standard stars. Theextinction parameters were fixed to the values given in the SOFIand SUSI-2 calibration tables . The magnitudes, airmass, zero-point and extinction parameters are reported in Table 1. Imagesin VRI JHK s are shown in Fig. 1. The NIR surface brightnessprofiles are shown in Fig. 2.The NIR spectra were reduced using the IRAF noao.twodspec package. Each individual image was cor-rected using standard procedures for the crosstalk and theflat-fielding. Consecutive images with the same jitter were com-bined together to obtain 4 output images. They were correctedfor the NIR sky, subtracting images with di ff erent jitter. Weextracted the source spectrum from the 4 images. We performedthe wavelength calibrations using a xenon arc extracted with thesame parameters as for the source. The 4 spectra were combinedin a single spectrum and corrected from telluric features usingthe spectrum of the standard star and the IRAF tool telluric .We applied this strategy for both blue and red filters and wefinally combined the blue and red spectra. We multiplied thefull spectrum by a calibrated spectrum of an A0 V (or G2 V)star from the spectral library of Pickles (1998) to avoid spectralcontamination from the telluric star. The output spectra weredereddened from the Galactic extinction using E (B − V) = .
3. Results and discussion
Only one optical and NIR candidate was located within the
Chandra error circle (see Fig. 1). Our NIR astrometry con-firmed the candidate 2MASS J09023731 − − / south-westwith a semi-major axis of 4 . (cid:48)(cid:48) (in K S ). The source was de-tected from filter R to filter K S . The source appears point-like inthe optical images. Since the X-ray observation was performedwith Chandra that has an angular resolution similar to the opti-cal / NIR images, it implies that we observed two di ff erent emit-ting components of the system between the X-rays / optical andthe NIR. The most obvious candidate is thus a galaxy with anactive galactic nucleus (AGN), plausibly a Seyfert galaxy, wherethe X-ray / optical radiation mainly comes from the accreting gassurrounding the supermassive black hole and the NIR radiationis mainly due to the dust present in the host galaxy (Wilkes2004). The NIR also suggests the presence of the spiral armsof the galaxy. The NIR magnitudes (reported in Table 1) do notshow any variation between the two observations separated by 5 retrieved at Table 1.
NIR and optical photometry of the single counterpart.
Date 2007–03–05T01:27–02:00Filters
J H K S Mag. (cid:5) . ± . . ± .
03 12 . ± . Z p . ± . . ± .
03 2 . ± . AM † (cid:48)(cid:48) ] 3.3 4.0 4.9P.A.[ ◦ ] − ± − ± − ± . ± .
09 0 . ± .
03 0 . ± . J H K S Mag. (cid:5) . ± .
02 13 . ± .
02 12 . ± . Z p . ± .
002 2 . ± .
02 2 . ± . AM † (cid:48)(cid:48) ] 3.3 4.0 4.9P.A.[ ◦ ] − ± − ± − ± . ± .
06 0 . ± .
03 0 . ± . V R I
Mag. > . ± . . ± . . ± . Z p − . ± . − . ± .
01 0 . ± . AM ext (cid:5) Total NIR magnitudes were extracted within ellipses centred on thesource and defined with the semi-major axis (SMA), the positionangle (P.A., relative to west axis and counter clockwise), and theeccentricity (Ecc.). † The NIR extinction parameters were fixed to
JHK S = . , . , .
03 with an error of ∆ ext / ext = . Fig. 2.
NIR surface brightness profiles of IGR J09026 − J − K = . ± .
08 (1 . ± .
13) mag for the March10th (March 5th) observation, after correcting for the Galacticextinction ( E (B − V) = . K − K s = − . × ( J − K ) (p.7, SOFI manual).Such strong reddening is mainly observed in Seyfert 1 galaxies(Rudy et al. 1982) and was interpreted as the reprocessing of hotdust located in the broad-line region (Barvainis 1987). Mid IRobservations are necessary to further study the dusty componentof the Seyfert galaxy. The NIR surface brightness profiles are in J.A. Zurita Heras et al.: Infrared identification of IGR J09026 − Table 2.
NIR spectroscopy. Emission lines are represented with the laboratory ( λ ) and fitted ( λ fit ) wavelengths. March 5 March 12Identification λ λ fit Flux EW FWHM λ fit Flux EW FWHM µ m µ m arbitrary Å Å µ m arbitrary Å Å[S III] 0.9531 – – – – 0.9911(1) 2724 ± − ± ± / Pa γ / ± − ±
11 222 ±
18 1.1293(2) 19513 ± − ± ± β ± − ± ± α ± − ± ±
21 1.9473(2) 12993 ± − ± ± Fig. 3.
NIR spectra of IGR J09026 − top ) andMarch 12 ( bottom ), 2007. The dashed lines represent the identi-fied emission lines red-shifted with z = . ± . S ν ∝ R / , seeFig. 2).The emission lines observed in the two spectra (see Fig. 3,the bottom spectrum having a higher signal-to-noise ratio) arereported in Table 2. The three main emission lines of the spec-tra situated at λ . µ m, λ . µ m and λ . µ m wereeasily identified with the HeI λ . µ m and HI (Pa β and Pa α )emission lines, once red-shifted by ∼ .
4. These are commonemission lines in AGN, independent of the class (see the spec-tral atlas of AGN in the energy range 0.8–2.4 µ m of Ri ff el et al.2006, which covers the spectral range of SOFI). These emissionlines are broad with values > − . Such broad emissionlines are typical of type 1 AGN.We also tried to identify other emission lines, particularlyin the spectrum taken on March 12, 2007 (see Fig. 3 bottom ):1) a narrow forbidden line [S III] λ . µ m with FWHM = ±
94 km s − , and 2) an HI Pa γ line might be blended withinthe HeI λ . µ m line, still the HeI line is always the strongestone in all AGN spectra (e.g. Fig. 9–12 in Ri ff el et al. 2006).Using only the HI Pa β and Pa α emission lines and both spectra,we derived an average redshift of z = . ± . α / Pa β gave F (Pa α ) / F (Pa β ) = . ± .
1. This value isslightly lower than the expected value of 1 . ± . . − × K and electron densities between 10 − cm − . Still, this is a common feature in Seyfertgalaxies (Glikman et al. 2006, and references therein).
4. Conclusion
Therefore, we conclude that IGR J09026 − z = . ± . INTEGRAL picked at z = .
033 (0 .
035 for IGRs). In the current census ofsources detected by
INTEGRAL / ISGRI, there are 165 detectedSeyfert galaxies (of which 63 are IGR sources). Discardingsources without reported redshift, the average redshift is 0.04(0.06 when considering only IGRs) with a variance of 0.005(0.006 for IGR) (Beckmann et al. 2009). IGR J09026 − INTEGRAL . Acknowledgements.
JAZH thanks Farid Rahoui, Volker Beckmann and SimonaSoldi for useful discussions on IGR J09026 − − − INTEGRAL . The authors acknowledge theuse of 1) NASA’s Astrophysics Data System, 2) the SIMBAD database, oper-ated at CDS, Strasbourg, France, and 3) data products from the Two MicronAll Sky Survey, which is a joint project of the University of Massachusetts andthe Infrared Processing and Analysis Center / California Institute of Technology,funded by the National Aeronautics and Space Administration and the NationalScience Foundation.
References
Barvainis, R. 1987, ApJ, 320, 537Beckmann, V., Soldi, S., Ricci, C., et al. 2009, subm. A&ABird, A. J., Barlow, E. J., Bassani, L., et al. 2006, ApJ, 636, 765Bird, A. J., Malizia, A., Bazzano, A., et al. 2007, ApJS, 170, 175Bodaghee, A., Courvoisier, T. J.-L., Rodriguez, J., et al. 2007, A&A, 467, 585Butler, S. C., Tomsick, J. A., Chaty, S., et al. 2009, ArXiv e-printsGlikman, E., Helfand, D. J., & White, R. L. 2006, ApJ, 640, 579Landolt, A. U. 1992, AJ, 104, 340Lebrun, F., Leray, J. P., Lavocat, P., et al. 2003, A&A, 411, L141Persson, S. E., Murphy, D. C., Krzeminski, W., Roth, M., & Rieke, M. J. 1998,AJ, 116, 2475Pickles, A. J. 1998, PASP, 110, 863Ri ff el, R., Rodr´ıguez-Ardila, A., & Pastoriza, M. G. 2006, A&A, 457, 61Rudy, R. J., Levan, P. D., & Rodriguez-Espinosa, J. M. 1982, AJ, 87, 598Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525Stephen, J. B., Bassani, L., Malizia, A., et al. 2006, A&A, 445, 869Tomsick, J. A., Chaty, S., Rodriguez, J., Walter, R., & Kaaret, P. 2008, ApJ, 685,1143Ubertini, P., Lebrun, F., Di Cocco, G., et al. 2003, A&A, 411, L131Wilkes, B. 2004, in Astronomical Society of the Pacific Conference Series, Vol.311, AGN Physics with the Sloan Digital Sky Survey, ed. G. T. Richards &P. B. Hall, 37 .A. Zurita Heras et al.: Infrared identification of IGR J09026 −4812 as a Seyfert 1 galaxy 5