NEAR-INFRARED SPECTROSCOPY OF SDSS J0303 – 0019: A LOW-LUMINOSITY, HIGH-EDDINGTON-RATIO QUASAR ATz∼ 6

Abstract

We present sensitive near-infrared (NIR) VLT ISAAC spectroscopic observations of the z = 6.08 quasar SDSS J030331.40-001912.9. This quasi-stellar object (QSO) is more than a magnitude fainter than other QSOs at z ∼ 6 for which NIR spectroscopy has been obtained to date and is therefore presumably more representative of the QSO population at the end of cosmic reionization. Combining rest-frame UV continuum luminosity with the width measurements of the Mg ii and C iv lines, we derive a black hole mass of 2 +1.0 −0.5 × 10 8 M , the lowest mass observed for z ∼ 6 QSOs to date, and derive an Eddington ratio of 1.6 +0.4 −0.6 , among the highest value derived for QSOs at any redshift. The Spitzer 24 μm nondetection of this QSO does not leave space for a significant hot dust component in its optical/NIR spectral energy distribution, in common with one other faint QSO at z = 6, but in contrast to more than 20 more z = 6 QSOs and all known lower redshift QSOs with sufficiently deep multiwavelength photometry. We conclude that we have found evidence for differences in the intrinsic properties of at least one z ∼ 6 QSO as compared to the lower redshift population.

EVIDENCE FOR NON-EVOLVING Fe II/Mg II RATIOS IN RAPIDLY ACCRETINGz∼ 6 QSOs

“…Notes. Column 1: QSO name; Columns 2 and 3: QSO right ascension and declination in J2000.0; Column 4: optical redshift from discovery paper; Column 5: NIR spectrum reference paper for the NIR spectra: A, Iwamuro et al (2002); B, Iwamuro et al (2004); C, Barth et al (2003); D, Kurk et al (2007); E, Kurk et al (2009);F, Jiang et al (2007).…”

“…The K-band observations were carried out with VLT-ISAAC with a spectral resolution of 450. We also included the K-band spectrum of J0303−0019 presented by Kurk et al (2009). This faint z ∼ 6 QSO was selected in SDSS Stripe 82 and its spectrum was taken with VLT-ISAAC.…”

“…This way, for each fit, we sample the γ parameter space with an average bin of 0.001 × 10 −17 erg s −1 cm −2 and we obtain ∼100 triplets satisfying the χ 2 , with a minimum of 15 (for the QSO BR 1033 − 0327). This condition is not satisfied for J0303 − 0019 (Kurk et al 2009; online Figure 9.8, top panels): in this case the Fe ii major features are severely affected by the telluric absorptions, resulting in a weaker constraint on the Fe ii template normalization which is in any case consistent with 0 (see Table 3). In Figure 2, we show two examples of the 3D χ 2 -cube projections and of the relative probability distribution for the Fe ii template normalization (similar plots for all sources discussed here are shown in Figure 9 and in online Figures 9.1-9.11).…”

“…Numerous NIR spectroscopy studies of QSO samples including z ∼ 6 sources have been carried out in the past (e.g., Maiolino et al 2001Maiolino et al , 2003Pentericci et al 2002;Iwamuro et al 2002Iwamuro et al , 2004Barth et al 2003;Dietrich et al 2003;Freudling et al 2003;Willott et al 2003;Jiang et al 2007;Kurk et al 2007Kurk et al , 2009. Their results show an increase in the scatter of the measured Fe ii/Mg ii line ratios as a function of the redshift.…”

“…However, the z ∼ 6 QSOs from SDSS only account for the brightest end of the QSO luminosity function. Only a few faint QSOs (z AB > 20) have measured M BH to date: two from the SDSS deep Stripe 82 (Jiang et al 2008(Jiang et al , 2009Kurk et al 2007Kurk et al , 2009), a deep imaging survey obtained by repeatedly scanning a stripe (260 deg 2 ) along the celestial equator, and nine from CFHQS (Willott et al 2010a). These sources are powered by less massive BHs (down to M BH ∼ 10 8 M ) that also appear to accrete close to the Eddington limit.…”