A study of quasars: an investigation into the molecular gas of a high-redshift quasar and the radio loudness of radio-quiet quasars
Student thesis: Doctoral Thesis
This thesis is composed of two parts; the first part deals with observations of the molecular gas towards an unlensed, obscured quasar AMS12, and the second part investigates radio undetected, optically selected quasi-stellar objects (QSOs) to determine the nature of the radio flux density distributions of these objects. AMS12 is an unlensed, obscured, z = 2.767 quasar which we observed with the Plateau de Bure Interferometer to detect carbon monoxide rotational transitions and atomic carbon fine structure lines in the molecular gas. We present new detections of the CO(5-4), CO(7-6), [CI]( ³P₁- ³P₀) and [CI](³P₂− ³P₁) molecular and atomic line transitions in this thesis. AMS12 is the first unlensed, high redshift source to have both atomic carbon ([CI]) transitions detected. The highly excited molecular gas probed by CO(3-2), (5-4) and (7-6), is modelled with large velocity gradient models. The gas kinetic temperature TG, density n(H₂), and the characteristic size r₀, are determined using the dust temperature from the far-infrared spectral energy distribution which had the following best-fitting parameters log₁₀[LFIR/L☉] = 13.5, dust temperature TD = 88 K and emissivity index β=0.6, as a prior for the gas temperature. The best fitting parameters are TG = 89.6 K, n(H₂) = 10 3.9 cm⁻³ and r₀ = 0.8 kpc. The ratio of the [CI] lines gives a [CI] excitation temperature of 43 ± 10 K, indicating the [CI] and the high-excitation CO are not in thermal equilibrium. The [CI] excitation temperature is below that of the dust temperature and the gas kinetic temperature of the high excitation CO, perhaps because [CI] lies at a larger radius where there may also be a large reservoir of CO at a cooler temperature, which may be detectable through the CO(1-0). Using the [CI]( ³P₁− ³P₀) line we can estimate the strength of the CO(1-0) line and hence the gas mass. This suggests that a significant fraction (~30%) of the molecular gas is missed from the high-excitation line analysis, giving a gas mass higher than that inferred from the assumption that the high-excitation gas is a good tracer of the low-excitation gas. The stellar mass was estimated from the mid-/near-infrared spectral energy distribution to be M* ~ 3 × 10¹¹M☉. The Eddington limited black hole mass is found from the bolometric luminosity to be M• ≳ 1.5×10⁹M☉. These give a black hole - bulge mass ratio of M•/M* ≳ 0.005. This is in agreement with studies on the evolution of the M•/M* relationship at high redshifts, which find a departure from the local value ~ 0.002. In the second half of the thesis we investigate the possible existence of a lower envelope in the radio luminosity versus optical luminosity plane. We select a population of QSOs from the Sloan Digital Sky Survey photometric quasar catalogue from Richards et al. The QSOs are within a narrow redshift band 0.3 < zphot < 0.5 and cross-matched with the 1.4 GHz National Radio Astronomy Observatory Very Large Array Sky Survey. The radio images extracted from the positions of the optical QSOs are retained if the flux integrated over the beam size of the radio survey is less than 3σIrms ≤ 1.35 mJy. The radio-undectected QSO population is split into eight samples depending on their optical magnitudes and stacked to determine the mean flux in each sample. The stacked mean flux is detected in all but the faintest optical magnitude sample. The radio versus optical luminosity relation from the stacked samples hint at a lower envelope in the radio luminosity which may be interpreted as there being a minimum radio jet power for a given accretion rate. Stacking assumes the underlying distribution of the property being measured is fairly represented by the stacked result. We investigate the underlying distribution of the radio flux density from the QSOs taking the noise of the sample into account. We find the distribution of the QSO flux density is modelled by a power-law with a negative index in all eight optical magnitude samples. This implies the mean stacked result is not a good representation of the distribution of the flux density of the QSOs and that there is no lower envelope. This highlights the danger of interpreting results from stacking without verifying the distribution is characterised by the mean stacked value. We appear to recover the quasar optical luminosity function when we model the distribution of radio loudness parameters suggesting that, since we are essentially holding the radio flux density fixed, the radio loudness is a function of the optical luminosity. This suggest that the radio loudness is not a fundamental property of the QSO but rather the ratio of two independent properties, the radio and optical luminosities. We convert the radio loudness parameter to jet efficiencies and find a minimum jet efficiency of ηmin = 4 × 10⁻⁴. We find there is no sign of a minimum jet efficiency as far as our data’s sensitivity limit allows, so we expect η<ηmin. Hence we provide an observational constraint for theoretical models of jet production in the minimum jet efficiency.
8 MB, PDF-document