The environment of weak emission-line quasars
aa r X i v : . [ a s t r o - ph . C O ] D ec Mon. Not. R. Astron. Soc. , 1–9 (2011) Printed 28 May 2018 (MN L A TEX style file v2.2)
The environment of Weak Emission-Line Quasars
M. Niko lajuk , ⋆ and R. Walter ⋆ ISDC Data Centre for Astrophysics, Chemin d’Ecogia 16, CH-1290 Versoix, Switzerland Faculty of Physics, University of Bialystok, Lipowa 41, PL-15424 Bia lystok, Poland
Accepted 2011 November 16; in original form 2011 July 26
ABSTRACT
The nature of weak emission-line quasars (WLQs) is probed by comparing theBaldwin effect (BEff) in WLQs and normal quasars (QSOs). We selected 81 high-redshift (z > L Bol /L Edd , and the X-ray to optical luminosity ratio, α ox , on theBEff is analysed. We find that WLQs follow a different relationship on the EW(C IV)- L Bol /L Edd plane than normal quasars. This relationship disagrees with the super-Eddington hypothesis. The weakness/absence of emission-lines in WLQs does notseem to be caused by their extremely soft ionizing continuum but by low coveringfactor (Ω / π ) of their broad line region (BLR). Comparing emission-line intensitiesindicates that the ratios of high-ionization line and low-ionization line regions (i.e.Ω HIL / Ω LIL ) are lower in WLQs than in normal QSOs. The covering factor of theregions producing C IV and Ly α emission-lines are similar in both WLQs and QSOs. Key words:
Galaxies: active – quasars: emission lines – Ultraviolet: galaxies – X-rays:galaxies.
A negative correlation between the broad emission-lineequivalent width (EW) and the luminosity in active galac-tic nuclei (AGNs) was discovered by Baldwin (1977) forthe C IV λ αλ λ λ λ λ ⋆ E-mail: [email protected], [email protected](MN); [email protected] (RW) more, an X-ray BEff in the iron K lines was detectedby Iwasawa & Taniguchi (1993) and analysed by e.g.Jiang, Wang & Wang (2006); Bianchi et al. (2007).Several physical explanations have been proposed to ex-plain the BEff. The most supported hypothesis is that themore luminous objects have softer UV/X-ray spectra reduc-ing ionization and photoelectric heating in the BLR gas. Ipsofacto the equivalent widths (EWs) of emission-lines are re-duced at higher luminosity with the strongest effect for high-ionization lines (HILs) (see Korista, Baldwin & Ferland1998 and Shields 2007 for a review of the BEff). Fundamen-tal parameters such as the Eddington ratio (Baskin & Laor2004; Warner, Hamann & Dietrich 2004; Bachev et al. 2004;Zhou & Wang 2005; Xu et al. 2008; Dong et al. 2009), theblack hole mass (Netzer, Laor & Gondhalekar 1992; Shields2007), or metalicity (Warner, Hamann & Dietrich 2004)have also been suggested as the driver of the BEff.The discovery of weak emission-line quasars (WLQs)i.e. sources with abnormally low broad emission-lines (e.g.EW(Ly α ) WLQ < . .
94) was discovered by McDowell et al. (1995). How- c (cid:13) M. Niko lajuk and R. Walter ever, up to 2009 only about 20 WLQs were known. Theymostly lie at high-redshifts (z > . .
62) and were found in the Sloan Digital Sky Sur-vey (SDSS) (Anderson et al. 2001; Schneider et al. 2003,2005; Collinge et al. 2005; Fan et al. 2006; Schneider et al.2007; Shemmer et al. 2009). Diamond-Stanic et al. (2009)recently discovered 65 new high-z WLQs, which may sug-gest that there is a deficit of weak line quasars below z <
2. However, Plotkin et al. (2010a,b) pointed out thatmore intermediate- and low-redshift WLQ may also exist.There is no generally accepted explanations for theweakness or even absence of emission-lines in WLQ. Sev-eral hypothesis were suggested by McDowell et al. (1995).Relativistic beaming in WLQ is not favoured as weak linequasars, in contrast to BL Lacs, are radio-quiet objects ,show no variability or strong polarization. Moreover, the ra-dio spectral slopes, connecting λ ∼ ∼
20 cm, aresignificantly steeper in radio-detected WLQs than the typi-cal slopes for BL Lac, α r ∼ . . Leighly et al. (2007a), based on observation of PHL1811, have claimed that its very soft spectral energy distri-bution (SED) (the photon index α ox = 2 . ± . is relatedto its super-Eddington nature (the estimated L Bol /L Edd liein the range 0.9-1.6). However, it is worth noting that, theobserved UV/optical part of the continuum in WLQs lookslike these of normal quasars. Richards et al. (2003) haveanalysed the spectra of 4576 SDSS quasars and they foundthat the spectral indices, α ν (where f ν ∝ ν α ν ), lie in awide range, with mean values from − .
25 to − .
76 (seetheir composites no. 1-4). The spectral indices in WLQs alsospan the same interval with a median values of α ν = − . The radio-loudness parameter R is defined as the ratio of therest-frame 6 cm to 2500 ˚A flux densities (see Jiang et al. 2007;Shen et al. 2011). Among 70 radio detected WLQs analysed byDiamond-Stanic et al. (2009) there is 81% of sources with R > “the very soft ionizing continuum” means that a continuum inthe far-UV (FUV) band is characterized by a steep spectrum.We use the X-ray to optical luminosity ratio, α ox (see definitionbelow). For typical quasar α ox is equal to − .
50 (Laor et al. 1997)and for the most luminous quasars with redshift 1.5-4.5 the mean α ox equals to -1.80 (Just et al. 2007). We adopt α ox < − . α ox = 0 . L ν (2keV) /L ν (2500˚A)] (e.g.Avni & Tananbaum 1982; Strateva et al. 2005;Gibson, Brandt & Schneider 2008). Wu et al. (2011) found a small population of X-ray weakquasars, suggesting that these PHL 1811 analogs possess theshielding gas with large covering factor. This gas absorbs al-most all soft X-ray continuum to prevent illumination ofbroad line region (BLR) by this radiation. As a result weakemission-lines are produced although a face-on observer seethe normal X-ray continuum.(2) The second hypothesis suggests that WLQs are nor-mal quasars with typical metallicities, ionizing continua,and ionization parameters, however, with an underdevel-oped BLR perhaps because of a freshly launched accretiondisc wind (Hryniewicz et al. 2010). The weakness/absenceof emission-lines in this case is caused by a low BLR cov-ering factor or a deficit of line-emitting gas in the BLR(Shemmer et al. 2010).In this paper we analyse both hypothesis: softness ofionizing continuum and underdevelopment of BLR. In sec-tion 2 we describe the sample of quasars that was used. Sec-tion 3 is devoted to the comparison of the observed proper-ties of WLQ and QSO that we discuss in section 4. The con-clusions are presented in section 5. We assume that H = 70km s − Mpc − , Ω m = 0 .
3, and Ω Λ = 0 . The sample of WLQs consists of 81 high-redshift (z > . L Bol = 46 . − . . ± . . ± . f /f = 45 .
3, Shen et al. 2011), (3) the UV continuumof SDSS J170109 can be fit as a power law ( f ν ∝ ν α ν ) witha spectral index α ν = − . ± . identical within errorsto that of the quasar composite from Richards et al. (2003).This value also differs from the mean spectral index calcu-lated for BL Lac candidates for which we have h α ν i = − . M BH , accretionrates in the Eddington units, L Bol /L Edd , and the spectralindices, α ox . Almost all but α ox values, were found in theSDSS DR7 quasar catalogue (Shen et al. 2011). Shen et al.(2011) point out that estimated EW(C IV) are encumbered This value is equivalent to α λ = − .
77 where f λ ∝ λ α λ .c (cid:13) , 1–9 he environment of WLQs binned SDSS J170109composite no.1 of Richards et al. Figure 1.
The rest-frame spectrum of SDSSJ170108.89+395443.0 binned and corrected for Galactic redden-ing using Cardelli, Clayton & Mathis (1989) relationship. Forcomparison, Richards et al. (2003) composite spectrum (no. 1) isshown. with large error when the signal-to-noise ratio (S/N) of theobserved WLQ spectrum is lower than 5 (see their figure 8).Therefore, in these cases we use EW values estimated byDiamond-Stanic et al. (2009) which for all sources but twohave EW(C IV) > σ . In other cases we use upper lim-its taken from the quasar catalogue or calculate them (seeTable 2). The spectral indices, α ox , of WLQs were takenfrom Shemmer et al. (2006, 2009). All those values originatefrom the Chandra observations. Additionally, we checked theChandra Multiwavelength Project Catalogue (Green et al.2009). We cross-correlated this catalogue with SDSS DR7.No new WLQs but SDSS J170109 were found. Its α ox isequal to − . Our aim is to compare the Baldwin effect observed in weakemission-line quasars to that observed in normal Type 1quasars. Fig. 2 displays the EW of the C IV emission-lineagainst the dimensionless accretion rates for different typesof quasars. This figure includes 81 quasars from the BrightQuasar Survey (BQS) with redshifts z < . L Bol = 44 . − . L Bol /L Edd , respectively. Dashed line represents thebest linear fit to their data (Baskin & Laor 2004, 2005).The triangles show 76 WLQs for which the EW and theaccretion ratios were calculated by Shen et al. (2011) orDiamond-Stanic et al. (2009). We must notice here that inboth Baskin & Laor’s and Shen et al.’s papers the methods to estimate L Bol /L Edd are similar. Both calculated the bolo-metric luminosity using relationship L Bol = BC λ × L λ, cont ,where L λ, cont is the continuum luminosity measured atwavelength λ and BC λ is the appropriate bolometric cor-rection factor. Both methods estimate the Eddington lumi-nosity, L Edd ∝ M BH using the scalling method in order tocalculate the black hole mass in AGN i.e. M BH ∝ L b λ, cont FWHM (ion). In this equation FWHM stands for the FullWidth at Half Maximum of ion which produces the emission-line. BQS quasars and high-z WLQs lie at different dis-tances, therefore, Baskin & Laor and Shen et al. used obser-vations of different emission-lines and continuum luminosi-ties to calculate M BH . Baskin & Laor (2004) used FWHMof H β line, L λ, cont measured at 5100 ˚A in the rest-frameof quasar and b = 0 .
50 (see equation (3) in Laor 1998).The authors used H β emission-line to estimate L Bol /L Edd because many scientists suggest non virialize character ofC IV (e.g. Risaliti, Young & Elvis 2009; Fine et al. 2010;Richards et al. 2011). However, high-z WLQs have redshiftshigher than 2.2. Therefore, Shen et al. (2011) used C IV lineand continuum luminosity observed at 1350˚A. They used therelationship determined by Vestergaard & Peterson (2006)between M BH , FWHM, and L λ, cont for which b = 0 . L Bol /L Edd . However, if one calculate Ed-dington ratios based on C IV emission-lines this relationshipis much weaker than the correlation with the L Bol /L Edd es-timated based on H β (Baskin & Laor 2005).In this paper we analyse 83 weak emission-line quasars,however, in Fig. 2 only 76 of them have C IV emission-line strong enough to determine their L Bol /L Edd (see Ta-ble 2). The EW of the remaining WLQs are lower than ∼ z = 0 . . ± . β emission-line isstrong with EW = 92 ˚A, whereas, O [III] λ λ . σ when for“genuine” WLQs is larger ( > σ ). We must mention, thatShemmer et al. (2009) decided to use EW(C IV) .
10 ˚A as ahallmark of WLQs. However, Diamond-Stanic et al. (2009)decided to use EW of Ly α +N V blend as those lines are bet-ter seen in distant weak line quasars. Therefore, we kept thisdefinition (i.e. EW(Ly α +N V) < . α +N V)in the sources with prominent C IV emission-lines is causedby strong absorption of the Ly α region and that they are,in fact, normal quasars.The best linear fit to BQS quasars sample (seen inFig. 2) suggests that WLQs follow a different relation-ship than normal quasars between EW and the accretionratio. The errors of those quantities for WLQ are largeand,unfortunately, we cannot fit a correlation to them. How-ever, in order to statistically quantify the hypothesis aboutdifferent relations we compare the reduced chi-squares, e χ ,in BQS’ and WLQ’s cases. We divide our WLQ objects into c (cid:13) , 1–9 M. Niko lajuk and R. Walter
WLQQSO (Baskin&Laor,04)
Figure 2.
Equivalent width of C IV measured at the rest frameplotted against accretion ratio. Filled blue squares show 81 BQSquasars analysed by Baskin & Laor (2004). Filled red trianglesand upper limits refer to 76 WLQs taken from Shen et al. (2011)or Diamond-Stanic et al. (2009). Points with error bars refer toobjects with a significance of EW higher than 5 σ . Otherwise,upper limits are shown. Dashed line is the best linear fit to BQSquasars (Baskin & Laor 2004, 2005). two subsets. The first one consists of all 76 weak emission-line quasars. In the second case we exclude all the upperlimits on EW(C IV) from our subset. We also assume thatthat the obtained fit parameters for normal quasars are alsothe same for WLQs. The estimated reduced chi-square is27.7 in the case of BQS quasars. That value is significantlylarger than 1. However, we must notice that there is a largespread in distribution of normal quasars around the linearfit. If we assume that a natural spread is less than 9 ˚A andwe calculate the fit avoiding outliers than the reduced chi-squares decreases to ≃ .
3. The estimated e χ for WLQs are ∼ ∼ L Bol /L Edd values are cal-culated using the method based on the luminosity- M BH re-lation, i.e. L Bol /L Edd ∝ FWHM(C IV) − . In many cases,the emission-lines in WLQ objects are broad and theirFWHM equals to a few thousand km s − (see Mg II inHryniewicz et al. 2010, H β in Shemmer et al. 2010 or C IVin Shen et al. 2011). Nevertheless, for weak C IV line (e.g.EW < a few ˚A) the FWHM value is underestimated thus L Bol /L Edd ratio is overestimated.The existence of normal accretion rates in WLQs wasdiscussed recently by Hryniewicz et al. (2010). They haveargued that when the Eddington ratio increases the widthof Ly α , Mg II lines decreases, C IV emission decreases,however, the Si IV line should become stronger, and theUV Fe II emission decreases. As the last two behaviours PHL1811TypicalerrorType 1 QSO (non-BAL)BAL QSOWLQ
Figure 3.
Rest frame equivalent width of C IV emission-line ver-sus spectral index α ox . Solid blue and open cyan squares refers toType 1 non-BAL and BAL quasars, respectively. Those 155 radio-quiet and radio-intermediate sources are taken from Green et al.(2009) paper. Solid red triangles and upper limits show 12 WLQobjects. Star denotes NLS1 PHL 1811. Solid line is the relation-ship obtained by Wu et al. (2009). Typical error of non-BAL andBAL QSOs is shown in the legend. are not observed in WLQ (see e.g. Schneider et al. 2010) ,Hryniewicz et al. (2010) claimed that the weakness of theemission-lines in WLQs is not caused by high L Bol /L Edd .So far, no observations of the FUV spectra of WLQswere made. Therefore, we analysed α ox which can shedlight on the shapes of the SED in the FUV/soft X-rayband (Fig. 3). Apart from the spectral indices for WLQswe analyzed together with them 155 normal quasars. Their α ox values were taken from the Chandra MultiwavelengthProject (Green et al. 2009). We cross-corellated this cata-logue with SDSS DR7 Quasar Catalogue (Shen et al. 2011)to obtain the EW(C IV) of quasars. The solid line in Fig. 3represents the best fit made by Wu et al. (2009). We mustnotice that this linear fit was made to another sample ofquasars, however, it fits very well to our sample of normalquasars. Our study clearly shows that α ox values in weakemission-line quasars span the same region as seen in non-BAL and BAL QSOs. It points out that the UV/soft X-ray SED of WLQs is similar to those seen in normal AGNsand proves that a soft ionizing continuum is not the rea-son for the weakness of the lines. That situation is foundin PHL 1811 which is NLS1 galaxy with super-Eddingtonrate ( L Bol /L Edd ∼ α ox = − .
3) (Leighly et al. 2007b). Therefore, PHL 1811follows the relationship estimated by Wu et al. (2009). c (cid:13) , 1–9 he environment of WLQs The WLQs are shifted vertically in the log EW(C IV)-log L Bol /L Edd plane relative to normal quasars (Fig. 2). Thisoffset and the fact that QSOs and WLQs SED are almostthe same, indicate that weak emission-line quasars are nor-mal AGNs, however, with intrinsically weak C IV emission-line. It is also clearly shown that the super-Eddington lu-minosities are not required in weak line quasars contrast-ing with the idea that WLQs are super-Eddington sources(Leighly et al. 2007b). Furthermore, the accretion rates inWLQs span the same interval as normal quasars (Fig. 2).The SEDs of weak line quasars observed in optical/UVband (till ∼ α ox of different quasars (Fig. 3). Sim-ilar analysis was carried out by Richards et al. (2011) orWu et al. (2011) (see their figure 9 or figure 6, respectively).We focus on weak emission-line quasars and enlarged oursample by adding objects with log EW(C IV) < .
6. Ouranalysis indicates that the UV/soft X-ray SED of WLQs issimilar to those of normal AGNs and a soft ionizing contin-uum is not the reason for the weakness of the lines.The intensity of an emission-line depends on the fluxof ionizing continuum, L ionise , and on the BLR gas coveringfactor, Ω / π : L (line) ∼ L ionise × Ω / π (see Ferland 2004,and his discussion for He II λ α ox ,measures by definition the ratio of the luminosities at 2 keVand at 2500˚A. If we assume that L ν (2500˚A) is roughly equalto L ν (1450˚A) and assuming that L ν (2keV) ≃ L ionise , we canwrite α ox ∼ log L ionise − log L ν (1450˚A). We can then expressthe line equivalent width as:log EW(line) ≈ const + log Ω4 π + α ox const The correlation EW(C IV)- α ox obtained for normal quasarsby e.g Wu et al. (2009) infers that the gas covering factorin BLR in Type 1 quasars is relatively constant. The gascovering factor in WLQ objects behaves differently (Fig. 3),suggesting that Ω WLQ is &
10 times smaller than in QSOs.Table 1 compares the emission-line intensity ratios ob-served in Seyfert 1.5 galaxy NGC 5548, normal quasars,and WLQs. We focus on 59 ‘real WLQs’, i.e. our selectedsubsample which consists of sources with EW(C IV) . α ) < α , C IV), intermediate-ionization lines (IILs; e.g. C III]), and low-ionization lines(LILs; such as Mg II, H β ). The C IV/Ly α intensity ratiofor different sources are the same from a statistical pointof view. The ratio of the covering factors of the regions re-sponsible for producing C IV and Ly α are therefore similarin WLQs and QSOs. Comparing low-, intermediate- withhigh-ionization lines we obtain that the ratios of the cover-ing factors of HIL/LIL and IIL/LIL are lower in WLQs thatin normal quasars. Even if for WLQ these ratios are basedon only few sources. This suggests that the covering factorof the BLR is smaller in WLQ. This is in agreement with observations of the weakH β emission-lines in SDSS J114153.34+021924.3 and SDSSJ123743.08+630144.9 (Shemmer et al. 2010). There authorshave explained the weakness of their emission-lines by adeficit of the BLR. The absence of BLR in WLQs have alsobeen recently suggested by Liu & Zhang (2011). The exis-tence of bright AGNs with dusty tori, but without BLRcould be understood when an anisotropic radiative pressureis released from an accretion disc. Liu & Zhang stated thatthis is possible just before the normal phase of an AGN.Additionally, Leighly & Moore (2004) suggested based onobservations of the emission-line profiles of NLS1 galaxiesIRAS 13224-3809 and 1H 0707-495 that the high-ionizationlines are produced in a wind and that the intermediate- andlow-ionization lines are produced in low-velocity gas associ-ated with the accretion disk at the base of the wind. Bothpictures are consistent with a suggestion that the regionsproducing emission-lines are developed by winds (Hawkins2004; Hryniewicz et al. 2010). In that case, when the BLR iscreated its covering factor is lower than estimated in normalAGNs.There is an observational analogy between weakemission-line quasars and the class of “optical dull” AGNs(also called XBONGs – “X-ray bright, optically nor-mal galaxies”). Their X-ray emission is bright while theylack both the broad emission-lines of Type 1 AGNsand the narrow emission-lines Type 2 AGNs (Elvis et al.1981; Comastri et al. 2002; Georgantopoulos & Georgakakis2005). There are a few hypothesis trying to answer the latentnature of XBONGs (see e.g. Moran, Filippenko & Chornock2002; Severgnini et al. 2003; Rigby et al. 2006; Civano et al.2007; Trump et al. 2009). However, none of them (such asdilution their spectra by a host galaxy, the low Eddingtonaccretion rate) can explain WLQs.Elvis (2000) has proposed an empirically derived struc-ture for quasars. He suggests presence of the funnel-shapedgeometrically thin accretion outflow which contains an highionized gas embedded in the colder BLR clouds. Accordingto our paper the low covering factor of the BLR means thatWLQ has got less clouds in the outflow or equivalently the“funnel” wind is geometrically thiner.Low covering factor of the BLR in WLQs would haveadditional consequence observed in the infrared (IR) band.Gaskell, Klimek & Nazarova (2007) have argued that thecovering factors of the BLR and of the dusty torus haveto be the same. It means that a small BLR in WLQs causesan evaporation of dust in the torus and a reduction ofits IR emissivity. Diamond-Stanic et al. (2009) mentionedthat two weak line quasars SDSS J140850.91+020522.7(with EW(C IV) = 1.95 ˚A) and SDSS J144231.72+011055.2(with EW(C IV) = 16.9 ˚A) are fainter in the IR ( ∼ µ m) band by 30-40%. Additionally, the IR flux densityof SDSS J130216.13+003032.1 (EW(C IV) = 27.8 ˚A) is alsorelatively low. More IR observations of WLQs are requiredto confirm this hypothesis. We have explored the Baldwin effect (BEff) in 82 high-redshift (z > c (cid:13) , 1–9 M. Niko lajuk and R. Walter
Table 1.
Arithmetic means and standard deviations of the emission-line intensity ratios. All observed line fluxes were dereddened forthe Milky Way contamination.Ratio NGC5548 PG QSO non-BAL QSO BAL QSO WLQ(all) WLQ(sub)(1) (2a) (2b) (3) (4) (5) (6) (7)C IV/Ly α . ± .
14 1 . ± .
57 0 . ± . a C IV/Mg II 5.80 4.98 4 . ± .
54 3 . ± .
47 3 . ± .
34 0 . ± . b C III]( λ . ± .
20 0 . ± . c Ly α /H β . ± .
25 1 . ± . d Column (1) refers the names of intensity ratios. In the case of Seyfert 1.5 galaxy NGC 5548 those values are shown in Column (2a) and(2b). In Column (2a) the Ly α , C IV, and C III] fluxes are taken from Korista et al. (1995), the Mg II flux fromGaskell, Klimek & Nazarova (2007), and the H β flux from Wanders & Peterson (1996). The ratios in Column (2b) are calculated fromcorrected for narrow-line fluxes and taken from Korista & Goad (2000). Column (3) refers to sample of 18 radio quiet PG quasars(Shang et al. 2007). Values of intensity ratio of radio-quiet and radio-intermediate 97 non-BAL and 14 BAL quasars are shown inColumns (4) and (5), respectively. Those sources was selected after cross-matching SDSS DR7 Quasar Catalogue (Shen et al. 2011)with Green et al. (2009) sample. This sample is consistent with sample used in the Fig. 3. In Column (6) we calculate mean ratio for allWLQs which show weak or strong C IV lines. Column (7) refers to subsample of WLQs, for which EW(C IV) <
20 ˚A and EW(Ly α ) < . a mean is calculated from 59 WLQs, b mean fromSDSS J094534 and SDSS J170109, c value only for SDSS J094534, d mean from SDSS J114153.34+021924.3 andSDSS J123743.08+630144.9. Data for intensity of C IV line in WLQs are taken from Shen et al. (2011), for Ly α fromDiamond-Stanic et al. (2009), for C IV/Mg II, and C III]/Mg II ratios from Hryniewicz et al. (2010), Hryniewicz et al. (in preparation),and for H β from Shemmer et al. (2010). quasars (WLQs) and compared them with a set of normalquasars. We draw the following conclusions: • The relationship between the rest-frame equivalentwidth for C IV emission-line and the Eddington ratio ob-served in WLQs has different normalization than for normalQSOs. This shift disagrees with the super-Eddington hy-pothesis (e.g. Shemmer et al. 2010). • The weakness or even the absence of emission-lines inWLQs is likely caused by a low covering factor of the broadline region (BLR) rather than by a very soft ionizing contin-uum. The comparison of the EW(C IV) and of the spectralindices, α ox , shows that the gas covering factor of the BLRin WLQs is &
10 times less than for normal QSOs. • The ratios of the covering factors of regions responsiblefor producing C IV and Ly α are similar in WLQs and QSOs. • The ratios of the covering factors Ω
HIL / Ω LIL are lowerin WLQ than in QSOs showing the deficit of the BLR inWLQ. However, this result is based on observations of onlyfour sources. • The radio-intermediate quasarSDSS J170108.89+395443.0 (z = 1 .
89) is a newintermediate-redshifted WLQ with rest-frame EW(C IV)= 2 . . • The definition of WLQ objects should take into accountnot only the weakness of Ly α or C IV emission-lines, sepa-rately, but both lines together. ‘False WLQs’ (sources withprominent C IV) are probably normal Type 1 quasars withintervening Ly α absorption. ACKNOWLEDGMENTS
We would like to thank an anonymous referee for useful com-ments that improve our paper. We are grateful to BozenaCzerny, Krzysztof Hryniewicz and Joanna Kuraszkiewicz foradvices during calculation and doing our analysis. We also thank Gary Ferland for pointing out a helpful article. MNthanks the Scientific Exchange Programme (Sciex) NMS ch for opportunity of working at the ISDC. This research hasbeen supported in part by the Polish MNiSW grants NN203380136, and 362/1/N-INTEGRAL/2008/09/0. REFERENCES
Anderson S. F. et al., 2001, AJ, 122, 503Avni Y., Tananbaum H., 1982, ApJL, 262, L17Bachev R., Marziani P., Sulentic J. W., Zamanov R., Cal-vani M., Dultzin-Hacyan D., 2004, ApJ, 617, 171Baldwin J. A., 1977, ApJ, 214, 679Baskin A., Laor A., 2004, MNRAS, 350, L31—, 2005, MNRAS, 356, 1029Bianchi S., Guainazzi M., Matt G., Fonseca Bonilla N.,2007, A&A, 467, L19Boroson T. A., Green R. F., 1992, ApJS, 80, 109Cardelli J. A., Clayton G. C., Mathis J. S., 1989, ApJ, 345,245Civano F. et al., 2007, A&A, 476, 1223Collinge M. J. et al., 2005, AJ, 129, 2542Comastri A. et al., 2002, ApJ, 571, 771Diamond-Stanic A. M. et al., 2009, ApJ, 699, 782Dietrich M., Hamann F., Shields J. C., Constantin A.,Vestergaard M., Chaffee F., Foltz C. B., JunkkarinenV. T., 2002, ApJ, 581, 912Dong X., Wang J., Wang T., Wang H., Fan X., Zhou H.,Yuan W., Long Q., 2009, in Astronomical Society of thePacific Conference Series, Vol. 408, The Starburst-AGNConnection, W. Wang, Z. Yang, Z. Luo, & Z. Chen, ed.,p. 83Elvis M., 2000, ApJ, 545, 63Elvis M., Schreier E. J., Tonry J., Davis M., Huchra J. P.,1981, ApJ, 246, 20Fan X. et al., 1999, ApJL, 526, L57—, 2006, AJ, 131, 1203 c (cid:13) , 1–9 he environment of WLQs Ferland G., 2004, in Astronomical Society of the PacificConference Series, Vol. 311, AGN Physics with the SloanDigital Sky Survey, G. T. Richards & P. B. Hall, ed., p.161Fine S., Croom S. M., Bland-Hawthorn J., Pimbblet K. A.,Ross N. P., Schneider D. P., Shanks T., 2010, MNRAS,409, 591Gaskell C. M., Klimek E. S., Nazarova L. S., 2007, ArXive-prints (arXiv:0711.1025)Georgantopoulos I., Georgakakis A., 2005, MNRAS, 358,131Gibson R. R., Brandt W. N., Schneider D. P., 2008, ApJ,685, 773Gilbert K. M., Peterson B. M., 2003, ApJ, 587, 123Green P. J. et al., 2009, ApJ, 690, 644Green P. J., Forster K., Kuraszkiewicz J., 2001, ApJ, 556,727Hawkins M. R. S., 2004, A&A, 424, 519Hryniewicz K., Czerny B., Niko lajuk M., Kuraszkiewicz J.,2010, MNRAS, 404, 2028Iwasawa K., Taniguchi Y., 1993, ApJL, 413, L15Jiang L., Fan X., Ivezi´c ˇZ., Richards G. T., Schneider D. P.,Strauss M. A., Kelly B. C., 2007, ApJ, 656, 680Jiang P., Wang J. X., Wang T. G., 2006, ApJ, 644, 725Just D. W., Brandt W. N., Shemmer O., Steffen A. T.,Schneider D. P., Chartas G., Garmire G. P., 2007, ApJ,665, 1004Keremedjiev M., Hao L., Charmandaris V., 2009, ApJ, 690,1105Kinney A. L., Rivolo A. R., Koratkar A. P., 1990, ApJ,357, 338Kong M., Wu X., Wang R., Liu F. K., Han J. L., 2006,A&A, 456, 473Korista K., Baldwin J., Ferland G., 1998, ApJ, 507, 24Korista K. T. et al., 1995, ApJS, 97, 285Korista K. T., Goad M. R., 2000, ApJ, 536, 284Kuraszkiewicz J. K., Green P. J., Forster K., Aldcroft T. L.,Evans I. N., Koratkar A., 2002, ApJS, 143, 257Laor A., 1998, ApJL, 505, L83Laor A., Fiore F., Elvis M., Wilkes B. J., McDowell J. C.,1997, ApJ, 477, 93Leighly K. M., 2004, ApJ, 611, 125Leighly K. M., Casebeer D., 2007, in Astronomical Societyof the Pacific Conference Series, Vol. 373, The CentralEngine of Active Galactic Nuclei, L. C. Ho & J.-W. Wang,ed., p. 365Leighly K. M., Halpern J. P., Jenkins E. B., Casebeer D.,2007a, ApJS, 173, 1Leighly K. M., Halpern J. P., Jenkins E. B., Grupe D.,Choi J., Prescott K. B., 2007b, ApJ, 663, 103Leighly K. M., Moore J. R., 2004, ApJ, 611, 107Liu Y., Zhang S. N., 2011, ApJL, 728, L44McDowell J. C., Canizares C., Elvis M., Lawrence A.,Markoff S., Mathur S., Wilkes B. J., 1995, ApJ, 450, 585Moran E. C., Filippenko A. V., Chornock R., 2002, ApJL,579, L71Netzer H., Laor A., Gondhalekar P. M., 1992, MNRAS,254, 15Plotkin R. M. et al., 2010a, AJ, 139, 390Plotkin R. M., Anderson S. F., Brandt W. N., Diamond-Stanic A. M., Fan X., MacLeod C. L., Schneider D. P.,Shemmer O., 2010b, ApJ, 721, 562 Pogge R. W., Peterson B. M., 1992, AJ, 103, 1084Richards G. T. et al., 2003, AJ, 126, 1131—, 2011, AJ, 141, 167Rigby J. R., Rieke G. H., Donley J. L., Alonso-Herrero A.,P´erez-Gonz´alez P. G., 2006, ApJ, 645, 115Risaliti G., Young M., Elvis M., 2009, ApJL, 700, L6Schneider D. P. et al., 2003, AJ, 126, 2579—, 2007, AJ, 134, 102—, 2005, AJ, 130, 367—, 2010, VizieR Online Data Catalog, 7260, 0Severgnini P. et al., 2003, A&A, 406, 483Shang Z., Wills B. J., Robinson E. L., Wills D., Laor A.,Xie B., Yuan J., 2003, ApJ, 586, 52Shang Z., Wills B. J., Wills D., Brotherton M. S., 2007,AJ, 134, 294Shemmer O., Brandt W. N., Anderson S. F., Diamond-Stanic A. M., Fan X., Richards G. T., Schneider D. P.,Strauss M. A., 2009, ApJ, 696, 580Shemmer O. et al., 2006, ApJ, 644, 86—, 2010, ApJL, 722, L152Shen Y. et al., 2011, ApJS, 194, 45Shields J. C., 2007, in Astronomical Society of the PacificConference Series, Vol. 373, The Central Engine of ActiveGalactic Nuclei, L. C. Ho & J.-W. Wang, ed., p. 355Strateva I. V., Brandt W. N., Schneider D. P., Vanden BerkD. G., Vignali C., 2005, AJ, 130, 387Trump J. R. et al., 2009, ApJ, 706, 797Vestergaard M., Peterson B. M., 2006, ApJ, 641, 689Wanders I., Peterson B. M., 1996, ApJ, 466, 174Warner C., Hamann F., Dietrich M., 2004, ApJ, 608, 136Wu J. et al., 2011, ApJ, 736, 28Wu J., Vanden Berk D. E., Brandt W. N., Schneider D. P.,Gibson R. R., Wu J., 2009, ApJ, 702, 767Xu Y., Bian W., Yuan Q., Huang K., 2008, MNRAS, 389,1703Zamorani G., Marano B., Mignoli M., Zitelli V., BoyleB. J., 1992, MNRAS, 256, 238Zhou X., Wang J., 2005, ApJL, 618, L83 c (cid:13) , 1–9 M. Niko lajuk and R. Walter
Table 2: The sample of weak emission-line quasars.Name z SDSS
EW(C IV) ref. log M BH α ox ref. log L Bol /L Edd (˚A) (M ⊙ )SDSS 010802.90-010946.1 3.330 . . ± . − . ± . . ± . . ± . − . ± . . − .
58 (3)SDSS 080523.32+214921.1 3.463 23 . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . ± . . ± . − . ± . . . ± .
101 0 . ± . . ± . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − .
49 (3) − . ± . . . ± .
956 0 . ± . . . ± . − . ± . . ± . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . ± . . ± . − . ± . . . ± .
069 0 . ± . . . ± .
167 0 . ± . . − .
54 (3)SDSS 114412.76+315800.8 3.235 6 . ± . . ± . − . ± . . . ± .
875 0 . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± .
775 0 . ± . . − .
93 (3)SDSS 121812.39+444544.5 4.518 . . ± . − . ± . . ± . . ± .
429 0 . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − . ± . . . ± . − .
37 (3) − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± . < − .
55 (3) 0 . ± . . ± . . ± .
029 0 . ± . . . ± . − . ± . . ± . . ± . − . ± . . . ± . − . ± . . − .
08 (3)SDSS 131429.00+494149.0 3.813 13 . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . c (cid:13) , 1–9 he environment of WLQs Table 2: continued.Name z SDSS
EW(C IV) ref. log M BH α ox ref. log L Bol /L Edd (˚A) (M ⊙ )SDSS 133146.20+483826.5 3.742 20 . ± . . ± . − . ± . . − .
70 (3)SDSS 134521.39+281822.2 4.082 . . ± . − . ± . . . ± . − . ± . . − .
54 (4)SDSS 141209.96+062406.9 4.466 . . ± . − . ± . . ± . . ± . − . ± . . . ± . − . ± . . . ± . − .
07 (3) − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± .
069 0 . ± . . − .
42 (4)SDSS 144803.36+240704.2 3.544 . . ± . − . ± . . . ± .
114 0 . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . . ± . − . ± . . . ± . − .
27 (5) − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . ± . . ± . − . ± . . . ± . − . ± . . . ± .
053 0 . ± . . ± . . ± . − . ± . . . ± . − . ± . α ox values were taken fromarticles with references shown in Columns (4) and (7). Numbers in parentheses correspond to the following references: (1) –Diamond-Stanic et al. (2009), (2) – this paper, (3) – Shemmer et al. (2009), (4) – Shemmer et al. (2006), (5) – Green et al.(2009). c (cid:13)000