We present an unprecedented spectroscopic survey of the CaII triplet + OI for a sample of 14 luminous ($-$26 $gtrsim$ M$_V$ $gtrsim$ $-$29), intermediate redshift (0.85 $lesssim$ $z$ $lesssim$ 1.65) quasars. The ISAAC spectrometer at ESO VLT allowed
us to cover the CaII NIR spectral region redshifted into the H and K windows. We describe in detail our data analysis which enabled us to detect CaII triplet emission in all 14 sources (with the possible exception of HE0048-2804) and to retrieve accurate line widths and fluxes of the triplet and OI $lambda$8446. The new measurements show trends consistent with previous lower $z$ observations, indicating that CaII and optical FeII emission are probably closely related. The ratio between the CaII triplet and the optical FeII blend at $lambda$4570 $AA$ is apparently systematically larger in our intermediate redshift sample relative to a low-$z$ control sample. Even if this result needs a larger sample for adequate interpretation, higher CaII/optical FeII should be associated with recent episodes of star formation in the intermediate redshift quasars and, at least in part, explain an apparent correlation of CaII triplet equivalent width with $z$ and $L$. The CaII triplet measures yield significant constraints on the emitting region density and ionization parameter, implying CaII triplet emission from log n$_H$ $gtrsim$ 11 [cm$^{-3}$] and ionization parameter log $U$ $lesssim$ 1.5. Line width and intensity ratios suggest properties consistent with emission from the outer part of a high density broad line region (a line emitting accretion disk?).
Black Hole Mass (M_BH) estimation in quasars, especially at high redshift, involves use of single epoch spectra with s/n and resolution that permit accurate measurement of the width of a broad line assumed to be a reliable virial estimator. Coupled w
ith an estimate of the radius of the broad line region this yields M_BH. The radius of the broad line region (BLR) may be inferred from an extrapolation of the correlation between source luminosity and reverberation derived r_BLR measures (the so-called Kaspi relation involving about 60 low z sources). We are exploring a different method for estimating r_BLR directly from inferred physical conditions in the BLR of each source. We report here on a comparison of r_BLR estimates that come from our method and from reverberation mapping. Our photoionization method employs diagnostic line intensity ratios in the rest-frame range 1400-2000 A (AlIII1860/SiIII]1892, CIV1549/AlIII1860) that enable derivation of the product of density and ionization parameter with the BLR distance derived from the definition of the ionization parameter. We find good agreement between our estimates of the density, ionization parameter and r_BLR and those from reverberation mapping. We suggest empirical corrections to improve the agreement between individual photoionization-derived r_BLR values and those obtained from reverberation mapping. The results in this paper can be exploited to estimate M_BH for large samples of high-z quasars using an appropriate virial broadening estimator. We show that the width of the UV intermediate emission lines are consistent with the width of H beta, therefore providing a reliable virial broadening estimator that can be measured in large samples of high-z quasars.
We describe a method for determination of physical conditions in the broad line regions of a significant subsample of Seyfert-1 nuclei and quasars. Several diagnostic ratios based on intermediate (AlIII 1860, SiIII 1892) and high (CIV 1549, SiIV 1397
) ionization lines in the UV spectra of quasars are used to constrain density, ionization and metallicity of the emitting gas. We apply the method to two extreme Population A quasars - the prototypical NLSy1 I Zw 1 and a high-z NLSy1-like object, SDSS J120144.36+011611.6. We find well-defined physical conditions: low ionization (ionization parameter $< 10^{-2}$), high density (10$^{12} - 10^{13}$ cm^{-3}) and significant metal enrichment. Ionization parameter and density can be derived independently for each source with an uncertainty that is always less than $pm 0.3$ in logarithm. We use the product density times ionization parameter to estimate the broad line region radius and the virial black hole mass. Estimates of black hole masses based on the photoionization analysis described in this paper are probably more accurate than those derived from the radius - luminosity correlation.
We present high S/N UV spectra for eight quasars at $zsim3$ obtained with VLT/FORS. The spectra enable us to analyze in detail the strongest emission features in the rest-frame range 1400-2000 AA of each source (ciii, siiii, aliii, siii, civ and siiv
). Previous work indicates that a component of these lines is emitted in a region with well-defined properties i.e., a high density and low ionization emitting region). Flux ratios aliii/siiii, civ/aliii, siiv/siiii, civ/siiv and siii/siiii for this region permit us to strongly constrain electron density, ionization parameter and metallicity through the use of diagnostic maps built from {sc CLOUDY} simulations. Reliable estimates of the product density times ionization parameter allow us to derive the radius of the broad line region rb from the definition of the ionization parameter. The rb estimate and the assumption of virialized motions in the line emitting gas yields an estimate for black hole mass. We compare our results with estimates obtained from the rb -- luminosity correlation customarily employed to estimate black hole masses of high redshift quasars.
