Fabrizio Brighenti

Hot Gas in and Around Elliptical Galaxies

▪ Abstract We review the origin, evolution, and physical nature of hot gas in elliptical galaxies and associated galaxy groups. Unanticipated recent X-ray observations with Chandra and XMM indicate much less cooling than previously expected. Consequently, many long-held assumptions must be reexamined or discarded and new approaches must be explored. Chief among these are the role of heating by active galactic nuclei, the influence of radio lobes on the hot gas, details of the cooling process, possible relation between the hot and colder gas in elliptical galaxies, and the complexities of stellar enrichment of the hot gas.

A Chandra View of Dark Matter in Early-Type Galaxies

We present a Chandra study of mass profiles in seven elliptical galaxies, of which three have galaxy-scale and four have group-scale halos, demarcated at 1013 M☉. These represent the best available data for nearby objects with comparable X-ray luminosities. We measure approximately flat mass-to-light (M/L) profiles within an optical half-light radius (Reff), rising by an order of magnitude at ~10 Reff, which confirms the presence of dark matter (DM). The data indicate hydrostatic equilibrium, which is also supported by agreement with studies of stellar kinematics in elliptical galaxies. The data are well fitted by a model comprising an NFW DM profile and a baryonic component following the optical light. The distribution of DM halo concentration parameters (c) versus Mvir agrees with ΛCDM predictions and our observations of bright groups. Concentrations are slightly higher than expected, which is most likely a selection effect. Omitting the stellar mass drastically increases c, possibly explaining large concentrations found by some past observers. The stellar M/LK agree with population synthesis models, assuming a Kroupa IMF. Allowing adiabatic compression (AC) of the DM halo by baryons made M/L more discrepant, casting some doubt on AC. Our best-fitting models imply total baryon fractions ~0.04-0.09, consistent with models of galaxy formation incorporating strong feedback. The groups exhibit positive temperature gradients, consistent with the universal profiles found in other groups and clusters, whereas the galaxies have negative gradients, suggesting a change in the evolutionary history of the systems around Mvir 1013 M☉.

The X-Ray Concentration-Virial Mass Relation

We present the concentration (c)-virial mass (M) relation of 39 galaxy systems ranging in mass from individual early-type galaxies up to the most massive galaxy clusters, (0.06-20) × 1014 M☉. We selected for analysis the most relaxed systems possessing the highest quality data currently available in the Chandra and XMM-Newton public data archives. A power-law model fitted to the X-ray c-M relation requires at high significance (6.6 σ) that c decreases with increasing M, which is a general feature of CDM models. The median and scatter of the c-M relation produced by the flat, concordance ΛCDM model (Ωm = 0.3, σ8 = 0.9) agrees with the X-ray data, provided that the sample is comprised of the most relaxed, early-forming systems, which is consistent with our selection criteria. When allowing only σ8 to vary in the concordance model, the c-M relation requires 0.76 99% confidence) both open CDM models and flat CDM models with Ωm ≈ 1. This result provides novel evidence for a flat, low-Ωm universe with dark energy using observations only in the local (z 1) universe. Possible systematic errors in the X-ray mass measurements of a magnitude ≈10% suggested by CDM simulations do not change our conclusions.

Probing the Dark Matter and Gas Fraction in Relaxed Galaxy Groups with X-Ray Observations from Chandra and XMM-Newton

We present radial mass profiles within ~0.3rvir for 16 relaxed galaxy groups—poor clusters (kT range 1-3 keV) selected for optimal mass constraints from the Chandra and XMM-Newton data archives. After accounting for the mass of hot gas, the resulting mass profiles are described well by a two-component model consisting of dark matter, represented by an NFW model, and stars from the central galaxy. The stellar component is required only for eight systems, for which reasonable stellar mass-to-light ratios (M/LK) are obtained, assuming a Kroupa IMF. Modifying the NFW dark matter halo by adiabatic contraction does not improve the fit and yields systematically lower M/LK. In contrast to previous results for massive clusters, we find that the NFW concentration parameter (cvir) for groups decreases with increasing Mvir and is inconsistent with no variation at the 3 σ level. The normalization and slope of the cvir-Mvir relation are consistent with the standard ΛCDM cosmological model with σ8 = 0.9 (considering a 10% bias for early forming systems). The small intrinsic scatter measured about the cvir-Mvir relation implies that the groups represent preferentially relaxed, early forming systems. The mean gas fraction (f = 0.05 ± 0.01) of the groups measured within an overdensity Δ = 2500 is lower than for hot, massive clusters, but the fractional scatter (σf/f = 0.2) for groups is larger, implying a greater impact of feedback processes on groups, as expected.

Three-dimensional simulations of the interstellar medium in dwarf galaxies - I. Ram pressure stripping

We present 3D hydrodynamic simulations of ram pressure stripping in dwarf galaxies. Analogous studies on this subject usually deal with much higher ram pressures, typical of galaxy clusters, or mild ram pressure due to the gas halo of the massive galactic companions. We extend over previous investigations by considering flattened, rotating dwarf galaxies subject to ram pressures typical of poor galaxy groups. We study the ram pressure effects as a function of several parameters such as galactic mass and velocity, ambient gas density, and angle between the galactic plane and the direction of motion. It turns out that this latter parameter plays a role only when the gas pressure in the galactic centre is comparable to the ram pressure. Despite the low values of the ram pressure, some dwarf galaxies can be completely stripped after 1-2 hundred of million years. This pose an interesting question on the aspect of the descents and, more in general, on the morphological evolution of dwarf galaxies. In cases in which the gas is not completely stripped, the propagation of possible galactic wind may be influenced by the disturbed distribution of the interstellar matter. We also consider the modification of the ISM surface density induced by the ram pessure and find that the resulting compression may trigger star formation over long time spans.

Far-Infrared Spitzer Observations of Elliptical Galaxies: Evidence for Extended Diffuse Dust

Far-infrared Spitzer observations of elliptical galaxies are inconsistent with simple steady state models of dust creation in red giant stars and destruction by grain sputtering in the hot interstellar gas at T ~ 107 K. The flux at 24 μm correlates with optical fluxes, suggesting that this relatively hot dust is largely circumstellar. But fluxes at 70 and 160 μm do not correlate with optical fluxes. Elliptical galaxies with similar LB have luminosities at 70 and 160 μm (L70 and L160) that vary over a factor of ~100, implying an additional source of dust unrelated to that produced by ongoing local stellar mass loss. Neither L70/LB nor L160/LB correlate with the stellar age or metallicity. Optical line fluxes from warm gas at T ~ 104 K correlate weakly with L70 and L160, suggesting that the dust may be responsible for cooling this gas. Many normal elliptical galaxies have emission at 70 μm that is extended to 5-10 kpc. Extended far-infrared emission with sputtering lifetimes of ~108 yr is difficult to maintain by mergers with gas-rich galaxies. Instead, we propose that this cold dust is buoyantly transported from reservoirs of dust in the galactic cores, which are supplied by mass loss from stars in the core. Intermittent energy outbursts from AGNs can drive the buoyant outflow.

Mechanical AGN feedback: controlling the thermodynamical evolution of elliptical galaxies

A fundamental gap in the current understanding of galaxies concerns the thermodynamical evolution of ordinary, baryonic matter. On the one hand, radiative emission drastically decreases the thermal energy content of the interstellar plasma (ISM), inducing a slow cooling flow towards the centre. On the other hand, the active galactic nucleus (AGN) struggles to prevent the runaway cooling catastrophe, injecting huge amount of energy into the ISM. The present study intends to investigate thoroughly the role of mechanical AGN feedback in (isolated or massive) elliptical galaxies, extending and completing the mass range of tested cosmic environments. Our previously successful feedback models in galaxy clusters and groups demonstrated that AGN outflows, self-regulated by cold gas accretion, are able to quench the cooling flow properly without destroying the cool core. Via three-dimensional hydrodynamic simulations (flash 3.3), also including stellar evolution, we show that massive mechanical AGN outflows can indeed solve the cooling-flow problem for the entire life of the galaxy, at the same time reproducing typical observational features and constraints such as buoyant underdense bubbles, elliptical shock cocoons, sonic ripples, dredge-up of metals, subsonic turbulence and extended filamentary or nuclear cold gas. In order to avoid overheating and totally emptying the isolated galaxy, the frequent mechanical AGN feedback should be less powerful and efficient (e∼ 10−4) compared with the heating required for more massive and bound ellipticals surrounded by the intragroup medium (e∼ 10−3).

XMM-NEWTON AND CHANDRA OBSERVATIONS OF THE GALAXY GROUP NGC 5044. I. EVIDENCE FOR LIMITED MULTIPHASE HOT GAS

Using new XMM and Chandra observations, we present an analysis of the temperature structure of the hot gas within a radius of 100 kpc of the bright nearby galaxy group NGC 5044. A spectral deprojection analysis of data extracted from circular annuli reveals that a two-temperature model (2T) of the hot gas is favored over single-phase or cooling flow ( = 4.5 ? 0.2 M? yr-1) models within the central ~30 kpc. Alternatively, the data can be fitted equally well if the temperature within each spherical shell varies continuously from ~Th to Tc ~ Th/2, but no lower. The high spatial resolution of the Chandra data allows us to determine that the temperature excursion Th ? Tc required in each shell exceeds the temperature range between the boundaries of the same shell in the best-fitting single-phase model. This is strong evidence for a multiphase gas having a limited temperature range. We do not find any evidence that azimuthal temperature variations within each annulus on the sky can account for the range in temperatures within each shell. We provide a detailed investigation of the systematic errors on the derived spectral models considering the effects of calibration, plasma codes, bandwidth, variable NH, and background rate. We find that the RGS gratings and the EPIC and ACIS CCDs give fully consistent results when the same models are fitted over the same energy ranges for each instrument. The cooler component of the 2T model has a temperature (Tc ~ 0.7 keV) similar to the kinetic temperature of the stars. The hot phase has a temperature (Th ~ 1.4 keV) characteristic of the virial temperature of the ~1013 M? halo expected in the NGC 5044 group. However, in view of the morphological disturbances and X-ray holes visible in the Chandra image within R ? 10 kpc, bubbles of gas heated to ~Th in this region may be formed by intermittent AGN feedback. Some additional heating at larger radii may be associated with the evolution of the cold front near R ~ 50 kpc, as suggested by the sharp edge in the EPIC images.

Feedback Heating in Cluster and Galactic Cooling Flows

Cluster cooling flow models that include both active galactic nuclei (AGN) heating and thermal conduction can reduce the overall mass cooling rate and simultaneously sustain density and temperature profiles similar to those observed. These computed flows have no ad hoc mass dropout. To achieve this agreement, the thermal conductivity must be about 0.35 ± 0.10 times the Spitzer value, similar to that advocated by Narayan & Medvedev. However, when applied to galaxy/group scales, the synergistic combination of AGN heating and conduction is less satisfactory. When the computed density profile and the global cooling rate are lowered by AGN heating to match observations of these smaller scale flows, the gas temperatures within ~10 kpc are too large. In addition, best-fitting flows in galaxy/groups with AGN heating and thermal conduction require conductivities much closer to the Spitzer value, ~0.5-1. Another difficulty with galaxy/group flows that combine AGN heating and conduction is that the iron enrichment by Type Ia supernovae is more effective when the gas density is lowered by heating to match the observations. The hot-gas iron abundance in galactic flows with heating and conduction greatly exceeds observed values throughout most of the galaxy. Galactic/group flows with central heating and conduction, therefore, require an additional process that removes the iron: failure of Type Ia supernovae ejecta to go into the hot phase, selective cooling, etc.

Star formation feedback and metal enrichment by Types Ia and II supernovae in dwarf spheroidal galaxies: the case of Draco

We present 3D hydrodynamic simulations aimed at studying the dynamical and chemical evolution of the interstellar medium in dwarf spheroidal galaxies. This evolution is driven by the explosions of Type II supernovae (SNe II) and Type Ia supernovae (SNe Ia), whose different contribution is explicitly taken into account in our models. We compare our results with detailed observations of the Draco galaxy. We assume star formation histories consisting . of a number of instantaneous bursts separated by quiescent periods. Diverse histories differ by the number of bursts, but all have the same total duration and give rise to the same amount of stars. Because of the large effectiveness of the radiative losses and the extended dark matter halo, no galactic wind develops, despite the total energy released by the supernovae is much larger than the binding energy of the gas. This explains why the galaxy is able to form stars for a long period (>3 Gyr), consistently with observations. In this picture, the end of the star formation and gas removal must result from external mechanisms, such as ram pressure and/or tidal interaction with the Galaxy. The stellar [Fe/H] distributions found in our models match very well the observed ones. We find a mean value ([Fe/H]) = -1.65 with a spread of ∼1.5 dex. The chemical properties of the stars derive by the different temporal evolution between SNe la and SNe II rate, and by the different mixing of the metals produced by the two types of supernovae. We reproduce successfully the observed [O/Fe]-[Fe/H] diagram. However, our interpretation of this diagram differs from that generally adopted by previous chemical models. In fact, we find that the break observed in the diagram is not connected with the onset of a galactic wind or with a characteristic time-scale for the sudden switchover of the SNe Ia, as sometimes claimed. Instead, we find that the chemical properties of the stars derive, besides the different temporal evolution of the SNe II and SNe Ia rates, from the spatial inhomogeneous chemical enrichment due to the different dynamical behaviour between the remnants of the two types of supernovae.