Partial coverage of the broad-line region of Q1232+082 by an intervening H2-bearing cloud*

Mon. Not. R. Astron. Soc. Partial coverage of the broad-line region of Q1232+082 by an intervening H2-bearing cloud S. A. Balashev 1 2 P. Petitjean 0 A. V. Ivanchik 1 2 C. Ledoux 4 R. Srianand 3 P. Noterdaeme 0 D. A. Varshalovich 1 2 0 Universite Pierre et Marie-Curie, Institut d'Astrophysique de Paris, CNRS-UMR7095 , 98bis bd Arago, 75014 Paris , France 1 St Petersburg State Polytechnical University , Polyteknicheskaya 29, 195251 St Petersburg , Russia 2 Ioffe Physical-Technical Institute of RAS , Polyteknicheskaya 26, 194021 St Petersburg , Russia 3 IUCAA , Post Bag 4, Ganesh Khind, Pune 411 007 , India 4 European Southern Observatory, Alonso de Co rdova 3107 , Casilla 19001, Vitacura, Santiago 19 , Chile A B S T R A C T We present a detailed analysis of the partial coverage of the Q1232+082 (zem = 2.57) broadline region (BLR) by an intervening H2-bearing cloud at zabs = 2.3377. Using curve of growth analysis and line profile fitting, we demonstrate that the H2-bearing component of the cloud covers the quasi-stellar object (QSO) intrinsic continuum source completely but only part of the BLR. We find that only 48 6 per cent of the C IV BLR emission is covered by the C I absorbing gas. We observe residual light (6 per cent) as well in the bottom of the O I 1302 absorption from the cloud, redshifted on top of the QSO Lyman emission line. Therefore, the extent of the neutral phase of the absorbing cloud is not large enough to cover all of the background source. The most likely explanation for this partial coverage is the small size of the intervening cloud, which is comparable to the BLR size. We estimate the number densities in the cloud: nH2 110 cm3 for the H2-bearing core and nH 30 cm3 for the neutral envelope. Given the column densities, N(H2) = 3.71 0.97 1019 cm2 and N(H I) = 7.94 1.6 1020 cm2, we derive the linear size of the H2-bearing core and the neutral envelope along the line of sight to be lH2 0.15+00..0055 pc and lH I 8.2+64..51 pc, respectively. We estimate the size of the C IV BLR by two ways: (i) extrapolating size-luminosity relations derived from reverberation observations and (ii) assuming that the H2-bearing core and the BLR are spherical in shape and the results are 0.26 and 0.18 pc, respectively. The large size we derive for the extent of the neutral phase of the absorbing cloud together with a covering factor of 0.94 of the Lyman emission means that the Lyman BLR is probably fully covered but that the Lyman emission extends well beyond the limits of the BLR. ISM; clouds - quasars; individual; Q1232+082 - cosmology; observations - The broad emission lines in the spectra of active galactic nuclei (AGN) respond to variations in the luminosity of the central continuum source with a delay due to light-traveltime effects within the emission-line region. It is therefore possible through the process of reverberation mapping to determine the geometry and kinematics of the emission-line region by careful monitoring of the continuum variations and the resulting emission-line response (Blandford & McKee 1982; Peterson 1993; Netzer & Peterson 1997). In particular the size of the broad-line region (BLR) can be inferred from the time delay measurement. Recent investigations of low-redshift AGN show a tight relation between this size and the luminosity of the AGN, R = A(L/1043)B, where R is the radius of the BLR, A is a typical distance in light-days and L is the H luminosity in erg s1. The index is found to have a value close to B 0.60.7 when the typical distance A is in the range 2080 light-days (Wu et al. 2004; Kaspi et al. 2005). Extending this relation to high luminosities yields a typical radius of the order of 1 pc for the BLR of bright high-z quasars. The size of the BLR has also been shown to be correlated with the luminosity in the continuum (Bentz et al. 2009). The anticorrelation found between the radius of the region over which an emission line is emitted and the velocity width of the broad emission line in the same object supports the idea that the BLR gas is virialized and its velocity field is dominated by the gravity of the central black hole (Peterson & Wandel 1999). If this is the case, then the BLR size and the emission line width give an estimate of the mass of the central object (Peterson & Wandel 1999; Warner, Hamann & Dietrich 2003; Wang et al. 2009). The BLR is stratified and the BLR reverberation mapping size for C IV is about half that for H. This is consistent with the above assumption as more highly ionized species are expected to be found primarily closer to the central source of ionization radiation. The spatial extent of the BLR is revealed by the partial coverage of some absorbing clouds, usually associated with the AGN, located in front of the quasar and producing absorption lines that are saturated but do not go to the zero flux level. Usually, the continuum source is covered completely but the emission-line region can be covered only partially (e.g. Petitjean, Rauch & Carswell 1994; Hamann 1997; Srianand & Shankaranarayanan 1999). In Wampler, Chugai & Petitjean (1995), four Fe II clouds are seen at different velocities with the similar covering factor, f 0.5. In Srianand et al. (2002), line locking and covering factors are shown to be intimately related and are used to constrain the geometry of the BLR. Covering factor is one of the characteristics together with variability and high metallicity that are used to distinguish intrinsic from intervening absorption systems. Indeed, partial coverage of intervening systems has rarely been reported. It has been the case in the early Keck spectrum of APM 08279+5255 (Ellison et al. 1999; Petitjean et al. 2000a) which is a lensed quasar whose images are separated by only 0.35 arcsec so that the Keck spectrum encompasses all the images (Ledoux et al. 1998). It is the case that the intervening Mg II systems are not covering all the lensed images (Lewis et al. 2002; Ellison et al. 2004) and, because of this, typical dimensions of the intervening clouds are derived to be of the order of 1 kpc. Partial coverage of a BLR by an intervening absorber had never been reported before Ivanchik et al. (2010). These authors note that the C I lines associated with the zabs = 2.3377 damped Lyman (DLA) system towards Q1232+082 probably do not cover the C IV BLR completely so that some flux stays unabsorbed at the bottom of saturated lines. In the present paper, we analyse in details this unique effect and test different interpretations. We present the observations in Section 2. Partial coverage is ascertained in Section 3. Physical conditions of the gas in the DLA are derived in Section 4 in order to infer its extent. Results are discussed in Section 5 before conclusions are drawn in Section 6. 2 O B S E RVAT I O N S The high-resolution spectrum of the high-redshift quasar Q1232+082 (zem = 2.57 and mV = 18.4) was obtained over several observing runs in the course of a survey for (...truncated)