We examine the black hole mass-galaxy bulge relationship in high-redshift QSOs. Black hole masses are derived from broad emission lines, and the host galaxy stellar velocity dispersion σ* is estimated from the widths of the radio CO emission lines. At redshifts z > 3, the CO line widths are narrower than expected for the black hole mass, indicating that these giant black holes reside in undersized bulges by an order of magnitude or more. The largest black holes (MBH > 109 M☉) evidently grow rapidly in the early universe without commensurate growth of their host galaxies. CO line widths offer a unique opportunity to study AGN host galaxy dynamics at high redshift.
Galaxy bulge luminosity L, black hole mass MBH, and stellar velocity dispersion σ* increase together in a way suggesting a close evolutionary relationship. Measurements of the MBH-σ* relationship as a function of cosmic time may shed light on the origin of this relationship. Direct measurements of σ* at high redshift are difficult, and the width of the narrow emission lines of active galactic nuclei (AGN) has been proposed as a surrogate for σ*. We investigate the utility of using σ for σ* by examining host galaxy magnitudes and [O III] line widths for low-redshift QSOs. For radio-quiet QSOs, σ is consistent in the mean with the value of σ* predicted by the Faber-Jackson relation. For our limited range of Lhost, scatter obscures the expected increase of σ with Lhost. However, for a sample of AGN covering a wide range of measured or inferred σ*, there is a clear increase of σ with σ*. Radio-loud QSOs on average have σ smaller by 0.1 dex than radio-quiet QSOs of similar Lhost, at least for luminosities typical of PG QSOs. Star formation rates in our low-redshift QSOs are smaller than required in order to maintain the typical observed ratio of bulge mass to black hole mass.
Supermassive black holes in galactic nuclei show a close relationship between the black hole mass, luminosity, and stellar velocity dispersion (sigma_*) of the host galaxy bulge. Probing these relationships at high redshift may shed light on the link between the formation of the galactic bulge and central black hole, but direct measurements of stellar velocity dispersion at high redshift are difficult. We show that [OIII] line widths provide a useful surrogate for sigma_* by comparing sigma_[OIII] with the value of sigma_* predicted by the Faber-Jackson relation for QSOs with measured host galaxy luminosity. Over a wide range of AGN luminosity, sigma_[OIII] tracks sigma_*, albeit with considerable scatter. [OIII] line widths are narrower by ~0.1 dex in radio-loud QSOs than in radio-quiet QSOs of similar host luminosity. In low redshift QSOs, the ratio of star formation rate to black hole growth rate is much smaller than the typical ratio of bulge mass to black hole mass.
The presence of supermassive black holes in galactic nuclei, both quiescent and active, is now well established. Advances in understanding the nature of the broad emission-line region make it possible to measure black hole masses in large numbers of active galactic nuclei (AGN). This in turn allows study of AGN properties as a function of black hole mass and Eddington ratio. Such studies should lead to a better understanding of the physics of AGN. The ability to measure black hole masses in QSOs also allows study of the relationship between black hole mass and host galaxy properties. Early results suggest that this relationship in QSOs at redshifts z approximate to 2 was similar to that in nearby galaxies today.
We use quasi-stellar object (QSO) emission-line widths to examine the MBH* relationship as a function of redshift and to extend the relationship to larger masses. Supermassive black holes in galactic nuclei are closely related to the bulge of the host galaxy. The mass of the black holeMBH increases with the bulge luminosity and with the velocity dispersion of the bulge stars, *. An important clue to the origin of this correlation would be an observational determination of the evolution, if any, in the MBH* relationship as a function of cosmic time. The high luminosity of QSOs affords the potential for studies at large redshifts. We derive black hole masses from the continuum luminosity and the width of the broad H line and * from the width of the narrow [O iii] lines. We find that radio-quiet QSOs conform to the establishedMBH* relationship up to values ofMBH 1010 M , with no discernible change in the relationship out to redshifts of z 3. These results are consistent with the idea that the growth of supermassive black holes and massive bulges occurred simultaneously. Subject headings: black hole physics — galaxies: active — quasars: general On-line material: color figures
LBQS 0103-2753 is a binary quasar with a separation of only 0.3 arcsec. The projected spacing of 2.3 kpc at the distance of the source (z = 0.848) is much smaller than that of any other known binary QSO. The binary nature is demonstrated by the very different spectra of the two components and the low probability of a chance pairing. LBQS 0103-2753 presumably is a galaxy merger with a small physical separation between the two supermassive black holes. Such objects may provide important constraints on the evolution of binary black holes and the fueling of AGN.
Imaging and spectroscopy with HST show that LBQS 0103-2753 (V = 17.8, z = 0.848) is a binary quasar with a separation of 0.3 arcsec or 2.3 kpc. This is by far the smallest separation binary quasar reported to date. The two components have very different spectra, including the presence of strong broad absorption lines (BALs) in component A only. The emission-line redshifts, based on the broad high ionization C IV lines, are z_A = 0.834 and z_B = 0.858; their difference is 3900 km/s in velocity units. The broad C IV lines, however, are probably not a good indicator of systemic redshift; and LBQS 0103-2753 A and B could have a much smaller systemic redshift difference, like the other known binary quasars. If the systemic redshift difference is small, then LBQS 0103-2753 would most likely be a galaxy merger that has led to a binary supermassive black hole. There is now one known 0.3 arcsec binary among roughly 500 QSOs that have been observed in a way that would reveal such a close binary. This suggests that QSO activity is substantially more likely for black hole binaries at spacings ~2 kpc than at ~15 to 60 kpc. Between 1987 and 1998, the observed Mg II BAL disappeared.
Some QSOs show an abrupt, strong rise in polarization at rest wavelength ∼750 Å. If this arises in the atmosphere of an accretion disk around a supermassive black hole, it may have diagnostic value. In PG 1222+228, the polarization rise occurs at the wavelength of a sharp drop in flux. We examine and reject interpretations of this feature involving a high‐velocity outflow. The observations agree with a model involving several intervening Lyman limit systems, two of which happen to coincide with the Lyman continuum polarization rise. After correction for the Lyman limit absorption, the continuum shortward of 912 Å is consistent with a typical power‐law slope, α ≈ -1.8. This violates the apparent pattern for the Lyman limit polarization rises to occur only in “candidate Lyman edge QSOs.” The corrected, polarized flux rises strongly at the wavelength of the polarization rise, resembling the case of PG 1630+377. The rise in polarized flux places especially stringent requirements on models.
We present measurements of the gas-phase abundance ratio C/O in six H II regions in the spiral galaxies M101 and NGC 2403, based on ultraviolet spectroscopy using the Faint Object Spectrograph on the Hubble Space Telescope. The ratios of C to O increase systematically with O/H in both galaxies, from log C/O≈-0.8 at log O/H=-4.0 to log C/O≈-0.1 at log O/H=-3.4. C/N shows no correlation with O/H. The rate of increase of C/O is somewhat uncertain because of uncertainty as to the appropriate UV reddening law and uncertainty in the metallicity dependence on grain depletions. However, the trend of increasing C/O with O/H is clear, confirming and extending the trend in C/O indicated previously from observations of irregular galaxies. Our data indicate that the radial gradients in C/H across spiral galaxies are steeper than the gradients in O/H. Comparing the data to chemical-evolution models for spiral galaxies shows that models in which the massive star yields do not vary with metallicity predict radial C/O gradients that are much flatter than the observed gradients. The most likely hypothesis at present is that stellar winds in massive stars have an important effect on the yields and thus on the evolution of carbon and oxygen abundances. C-to-O and N-to-O abundance ratios in the outer disks of spirals determined to date are very similar to those in dwarf irregular galaxies. This implies that the outer disks of spirals have average stellar-population ages much younger than those of the inner disks.
Some QSOs show an abrupt increase in polarization at rest wavelengths below about 750 Angstrom. Blaes & Agol interpreted this in terms of stellar atmosphere effects in an accretion disk around a supermassive black hole. We have computed the locally emitted continuum energy distribution and polarization from a QSO disk and the resulting observed polarized spectrum including the general, relativistic transfer function. Relativity adds an additional blueshift to the wavelength of the polarization rise, relative to the rest wavelength of the Lyman edge (912 Angstrom), such that the model does not agree with the observations. A good fit results from a model in which the polarization jumps sharply at the Lyman edge in the rest frame of the orbiting gas. Relativistic effects give an abrupt, blueshifted rise in polarization, in agreement with the observations. This model offers a potential method for constraining the inclination of the disk and the angular momentum of the black hole.
We present new long-slit spectra of 12 H II regions in the Scd spiral galaxy NGC 2403. Gas-phase element abundances for O, N, S, Ne, and Ar were estimated using direct measurements of electron temperature based on detections of [O III] lambda 4363, [S III] lambda 6312, and [O II] lambda 7320-7330, and from theoretical photoionization analysis. We find abundance gradients for O/H and N/O of -0.102 +/- 0.009 dex kpc(-1) and -0.032 +/- 0.005 dex kpc(-1), respectively. The relatively flat N/O gradient suggest a significant source of primary nitrogen. An upper limit for the C/O ratio for one H II region was determined from a Hubble Space Telescope FOS spectrum. A mild outward increasing gradient in S/O is seen at marginal significance (0.03 +/- 0.02 dex kpc(-1)).We compare the abundance gradient and effective oxygen yield deduced for NGC 2403 with those determined for M33, another Scd galaxy with very similar structural parameters. We find close agreement in most of the chemical properties between NGC 2403 and M33. However, the effective yield for oxygen determined from closed box chemical evolution calculations is higher in M33 than in NGC 2403. The higher yield derived for M33 is similar to the case of H I-deficient Virgo spiral galaxies.We also compare NGC 2403 and M33 with a larger sample of unbarred spirals having abundance measurements. The global metallicity of the spirals correlates well with galaxy luminosity, as already noted from earlier investigations. The O/H gradient per unit disk scale length does not correlate with galaxy luminosity with possibly small intrinsic scatter, suggesting that spiral galaxies are homologous with regard to chemical evolution. The correlation between gas abundances and local surface brightness (mass density) in late-type spirals appears to be described well by chemical evolution models incorporating self-regulating star formation. However, the characteristic abundance at a given value of surface brightness correlates with galaxy luminosity. This suggests that an additional parameter which scales with galaxy mass influences the chemical properties of disks.
Novae have been proposed as the explanation of high reported abundances of heavy elements in the gas producing the broad absorption lines (BALs) of QSOs. High abundances of odd numbered elements, including aluminum, are predicted. Available data contains hints that the Al/Si ratio may be high both in the BAL gas and in the broad emission-line gas.
We present new measurements of the gas-phase C/O abundance ratio in both the northwest (NW) and southeast (SE) components of the extremely metal-poor dwarf irregular galaxy I Zw 18, based on ultraviolet spectroscopy of the two H II regions using the Faint Object Spectrograph on the Hubble Space Telescope. We determine values of log C/O = -0.63 ± 0.10 for the NW component and log C/O = -0.56 ± 0.09 for the SE component. In comparison, log C/O = -0.37 in the Sun, while log C/O = -0.85 ± 0.07 in the three most metal-poor irregular galaxies measured by Garnett et al. Our measurements show that C/O in I Zw 18 is significantly higher than in other comparably metal-poor irregular galaxies and above predictions for the expected C/O from massive star nucleosynthesis. These results suggest that carbon in I Zw 18 has been enhanced by an earlier population of lower-mass carbon-producing stars; this idea is supported by stellar photometry of I Zw 18 and its companion, which demonstrate that the current bursts of massive stars were not the first. Despite its very low metallicity, it is likely that I Zw 18 is not a “primeval” galaxy.
The broad absorption lines (BALs) of QSOs indicate abundances of heavy elements, relative to hydrogen, that are 1-2 orders of magnitude higher than the solar values. In at least one QSO, an especially large enhancement of phosphorus is observed. These abundances resemble those in Galactic novae, and this suggests that novae may produce the BAL gas. The needed rate of nova outbursts may come from single white dwarfs that accrete gas as they pass through a supermassive accretion disk around a central black hole.
We present new measurements of chemical abundances in H II regions in spiral galaxies of the Virgo cluster and a comparison of Virgo galaxies and field spirals. With these new data there now exist nine Virgo spirals with abundance measurements for at least four H II regions. Our sample of Virgo galaxies ranges from H I deficient objects near the core of the cluster to galaxies with normal H I properties, far from the cluster core. We investigate the relationship between H I disk characteristics and chemical abundances to determine whether dynamical process that remove gas from the disk, such as ram pressure stripping by the intracluster medium, also affect the chemical abundances.
We present UV observations of seven H II regions in low-luminosity dwarf irregular galaxies and the Magellanic Clouds obtained with the Faint Object Spectrograph on the Hubble Space Telescope (HST) in order to measure the C/O abundance ratio in the interstellar medium (ISM) of those galaxies. We measure both O III] 1666 Angstrom and C III] 1909 Angstrom in our spectra, enabling us to determine C+2/O+2 with relatively small uncertainties. The results from our HST observations show a continuous increase in C/O with increasing O/H, consistent with a power law having an index of 0.43 +/- 0.09 over the range -4.7 to -3.6 in log (O/H). One possible interpretation of this trend is that the most metal-poor galaxies are the youngest and dominated by the products of early enrichment by massive stars, while more metal-rich galaxies show increasing, delayed contributions of carbon from intermediate-mass stars. However, recent evolution models for massive stars including mass loss suggest that the yield of carbon from massive stars may increase with metallicity relative to the yield of oxygen; new chemical evolution models for the solar neighborhood which include nucleosynthesis from these recent stellar evolution models predict a C/O abundance evolution similar to that observed in the metal-poor galaxies. The trend in the C/N ratio increases steadily with O/H in the irregular galaxies, but decreases suddenly for solar neighborhood stars and H II regions. This may indicate that the bulk of nitrogen production is decoupled from the synthesis of carbon in our Galaxy. Our results also suggest that it may not be appropriate to combine abundances in irregular galaxies with those in spiral galaxies to study the evolution of chemical abundances. Our measured C/O ratios in the most metal-poor galaxies are consistent with predictions of nucleosynthesis from massive stars for Weaver SZ Woosley's best estimate for the C-12(alpha, gamma)O-16 nuclear reaction rate, assuming negligible contamination from carbon produced in intermediate-mass stars in these galaxies.We detect a weak N III] 1750 Angstrom multiplet in SMC N88A and obtain interesting upper limits for two other objects. Our 2 sigma upper limits on the 1750 Angstrom feature indicate that the N+2/O+2 ratios in these objects are not significantly larger than the N+/O+ ratios measured from optical spectra. This behavior is consistent with predictions of photoionization models, although better detections of N III] are needed to confirm the results.
We have measured the Si III] intercombination doublet at 1883 Å and 1892 Å in seven extragalactic H II regions using the Faint Object Spectrograph on the Hubble Space Telescope. These measurements were used to derive the Si/C and Si/O gas phase abundance ratios in the ionized gas. We find that Si/O shows no systematic variation with O/H over the range -4.8 < log O/H < -3.4. The weighted mean value for Si/O over this range is log Si/O = -1.59 ± 0.07. For comparison, the solar value is log Si/O = -1.37, while for Galactic B stars and supergiants -1.6 < log Si/O < -1.2. Uncertainty in the stellar reference value for Si/O prevents a straightforward interpretation of our Si/O ratios. However, if the solar Si/O ratio represents the intrinsic cosmic ratio, our results imply that, on average, only about 50% of the total abundance of silicon in these H II regions is incorporated into dust grains, a value significantly smaller than is typically measured in dense interstellar clouds. Our results suggest that some grain modification is occurring within the H II region environment.