arXiv subjects

D. John Hillier

Publications and source records attributed to D. John Hillier.

116 records · Page 7Linked to original sources

arXiv2005

Quantitative spectroscopic analysis of and distance to SN1999em

This work presents a detailed quantitative spectroscopic analysis of, and the determination of the distance to, the type II supernovae (SN) SN1999em with CMFGEN (Dessart & Hillier 2005a), based on spectrophotometric observations at eight dates up to 40 days after discovery. We use the same iron-group metal content for the ejecta, the same power-law density distribution (with exponent n~10), and a Hubble-velocity law at all times. We adopt a H/He/C/N/O abundance pattern compatible with CNO-cycle equilibrium values for a RSG/BSG progenitor, with C/O enhanced and N depleted at later times. Based on our synthetic fits to spectrophotometric observations of SN1999em, we obtain a distance of 11.5Mpc, similar to that of Baron et al. (2004) and the Cepheid distance to the galaxy host of 11.7Mpc (Leonard et al. 2003). Similarly, based on such models, the Expanding Photosphere Method (EPM) delivers a distance of 11.6Mpc, with negligible scatter between photometric bandpass sets; there is thus nothing wrong with the EPM as such. Previous determinations using the tabulated correction factors of Eastman et al. (1996) all led to 30-50% underestimates: we find that this is caused by 1) an underestimate of the correction factors compared to the only other study of the kind by Dessart & Hillier (2005b), 2) a neglect of the intrinsic >20% scatter of correction factors, and 3) the use of the EPM at late times when severe line blanketing makes the method inaccurate. The need of detailed model computations for reliable EPM distance estimates thus defeats the appeal and simplicity of the method. However, detailed fits to SN optical spectra, based on tailored models for individual SN observations, offers a promising approach to obtaining distances with 10-20% accuracy, either through the EPM or a la Baron et al. (2004).

astro-ph↗

arXiv2005

On the Steady Nature of Line-Driven Disk Winds: Application to Cataclysmic Variables

We apply the semi-analytical analysis of the steady nature of line-driven winds presented in two earlier papers to disk winds driven by the flux distribution of a standard Shakura & Sunyaev (1973) disk for typical cataclysmic variable (CV) parameters. We find that the wind critical point tends to be closer to the disk surface towards the inner disk regions. Our main conclusion, however, is that a line-driven wind, arising from a steady disk flux distribution of a standard Shakura-Sunyaev disk capable of locally supplying the corresponding mass flow, is steady. These results confirm the findings of an earlier paper that studied "simple" flux distributions that are more readily analyzable than those presented here. These results are consistent with the steady velocity nature of outflows observationally inferred for both CVs and quasi-stellar objects (QSOs). We find good agreement with the 2.5D CV disk wind models of Pereyra and collaborators. These results suggest that the likely scenario to account for the wind outflows commonly observed in CVs is the line-driven accretion disk wind scenario, as suggested early-on by Cordova & Mason (1982). For QSOs, these results show that the line-driven accretion disk wind continues to be a promising scenario to account for the outflows detected in broad absorption line (BAL) QSOs, as suggested early-on by Turnshek (1984), and analyzed in detail by Murray et al. (1995).

astro-ph↗

arXiv2005

Characteristic QSO Accretion Disk Temperatures from Spectroscopic Continuum Variability

Using Sloan Digital Sky Survey (SDSS) quasar spectra taken at multiple epochs, we find that the composite flux density differences in the rest frame wavelength range 1300-6000 AA can be fit by a standard thermal accretion disk model where the accretion rate has changed from one epoch to the next (without considering additional continuum emission components). The fit to the composite residual has two free parameters: a normalizing constant and the average characteristic temperature $\bar{T}^*$. In turn the characteristic temperature is dependent on the ratio of the mass accretion rate to the square of the black hole mass. We therefore conclude that most of the UV/optical variability may be due to processes involving the disk, and thus that a significant fraction of the UV/optical spectrum may come directly from the disk.

astro-ph↗

arXiv2005

Distance determinations using type II supernovae and the expanding photosphere method

Due to their high intrinsic brightness, type II supernovae (SN) can be used as lighthouses to constrain distances in the Universe using variants of the Baade-Wesselink method. Based on a large set of CMFGEN models (Hillier & Miller 1998) covering the photospheric phase of type II SN, we study the various concepts entering one such technique, the Expanding Photosphere Method (EPM). We compute correction factors $ξ$ needed to approximate the synthetic Spectral Energy Distribution (SED) with that of a blackbody at temperature $T$. Our $ξ$, although similar, are systematically greater, by $\sim 0.1$, than the values obtained by Eastman et al. (1996). This translate into a systematic enhancement of 10-20% in EPM-distances. We find that line emission and absorption, not directly linked to color temperature variations, can considerably alter the synthetic magnitude, and cause above-unity $ξ$-values. Optically-thick lines, present in the optical at late-times, can also introduce a strong wavelength-dependence of the photospheric radius, invalidating the use of the Baade method. Both the impact of line-blanketing on the SED and the photospheric radius at low $T$ suggest that the EPM is best used at early times, when the outflow is fully ionized and line-opacity mostly confined to the UV range. We also investigate how reliably one can measure the photospheric velocity from P-Cygni line profiles. Contrary to the usually held belief, the velocity at maximum absorption in the P-Cygni trough of optically-thick lines can both {\it overestimate or underestimate} the photospheric velocity, with a magnitude that depends on the SN outflow density gradient and the optical thickness of the line.

astro-ph↗

arXiv2004

Metallicity in the Galactic Center: The Arches cluster

We present a quantitative spectral analysis of five very massive stars in the Arches cluster, located near the Galactic center, to determine stellar parameters, stellar wind properties and, most importantly, metallicity content. The analysis uses a new technique, presented here for the first time, and uses line-blanketed NLTE wind/atmosphere models fit to high-resolution near-infrared spectra of late-type nitrogen-rich Wolf-Rayet stars and OfI+ stars in the cluster. It relies on the fact that massive stars reach a maximum nitrogen abundance that is related to initial metallicity when they are in the WNL phase. We determine the present-day nitrogen abundance of the WNL stars in the Arches cluster to be 1.6% (mass fraction) and constrain the stellar metallicity in the cluster to be solar. This result is invariant to assumptions about the mass-luminosity relationship, the mass-loss rates, and rotation speeds. In addition, from this analysis, we find the age of the Arches cluster to be 2-2.5Myr, assuming coeval formation.

astro-ph↗

arXiv2003

On the Steady Nature of Line-Driven Disk Winds

We perform an analytic investigation of the stability of line-driven disk winds, independent of hydrodynamic simulations. Our motive is to determine whether or not line-driven disk winds can account for the wide/broad UV resonance absorption lines seen in cataclysmic variables (CVs) and quasi-stellar objects (QSOs). In both CVs and QSOs observations generally indicate that the absorption arising in the outflowing winds has a steady velocity structure on time scales exceeding years (for CVs) and decades (for QSOs). However, published results from hydrodynamic simulations of line-driven disk winds are mixed, with some researchers claiming that the models are inherently unsteady, while other models produce steady winds. The analytic investigation presented here shows that if the accretion disk is steady, then the line-driven disk wind emanating from it can also be steady. In particular, we show that a gravitational force initially increasing along the wind streamline, which is characteristic of disk winds, does not imply an unsteady wind. The steady nature of line-driven disk winds is consistent with the 1D streamline disk-wind models of Murray and collaborators and the 2.5D time-dependent models of Pereyra and collaborators. This paper emphasizes the underlying physics behind the steady nature of line-driven disk winds using mathematically simple models that mimic the disk environment

astro-ph↗

arXiv2002

Stellar and Wind Properties of LMC WC4 stars - A metallicity dependence for Wolf-Rayet mass-loss rates

We use ultraviolet space-based (FUSE, HST) and optical/IR ground-based (2.3m MSSSO, NTT) spectroscopy to determine the physical parameters of six WC4-type Wolf-Rayet stars in the Large Magellanic Cloud. Stellar parameters are revised significantly relative to Grafener et al. (1998) based on improved observations and more sophisticated model atmosphere codes, which account for line blanketing and clumping. We find that stellar luminosities are revised upwards by up to 0.4 dex, with surface abundances spanning a lower range of 0.1<C/He<0.35 (20-45% carbon by mass) and O/He<0.06 (<10% oxygen by mass). Relative to Galactic WC5-8 stars at known distance, and analysed in a similar manner, LMC WC4 stars possess systematically higher stellar luminosities, 0.2dex lower wind densities, yet a similar range of surface chemistries. We illustrate how the classification CIII 5696 line is extremely sensitive to wind density, such that this is the principal difference between the subtype distribution of LMC and Galactic early-type WC stars. Temperature differences do play a role, but carbon abundance does not affect WC spectral types. We illustrate the effect of varying temperature and mass-loss rate on the WC spectral type for HD32257 (WC4, LMC) and HD156385 (WC7, Galaxy) which possess similar abundances and luminosities. Using the latest evolutionary models, pre-supernova stellar masses in the range 11-19 Mo are anticipated for LMC WC4 stars, with 7-14 Mo for Galactic WC stars with known distances. These values are consistent with pre-cursors of bright Type-Ic supernovae such as SN1998bw (alias GRB 980425) for which a minimum total mass of C and O of 14Mo has been independently derived.

astro-ph↗

arXiv2000

Quantitative analysis of WC stars: Constraints on neon abundances from ISO/SWS spectroscopy

Neon abundances are derived in four Galactic WC stars -- gamma Vel (WR11, WC8+O7.5III), HD156385 (WR90, WC7), HD192103 (WR135, WC8), and WR146 (WC5+O8) - using mid-infrared fine structure lines obtained with ISO/SWS. Stellar parameters for each star are derived using a non-LTE model atmospheric code (Hillier & Miller 1998) together with ultraviolet (IUE), optical (INT, AAT) and infrared (UKIRT, ISO) spectroscopy. In the case of gamma Vel, we adopt results from De Marco et al. (2000), who followed an identical approach. ISO/SWS datasets reveal the [NeIII] 15.5um line in each of our targets, while [NeII] 12.8um, [SIV] 10.5um and [SIII] 18.7um are observed solely in gamma Vel. Using a method updated from Barlow et al. (1988) to account for clumped winds, we derive Ne/He=3-4x10^-3 by number, plus S/He=6x10^-5 for gamma Vel. Neon is highly enriched, such that Ne/S in gamma Vel is eight times higher than cosmic values. However, observed Ne/He ratios are a factor of two times lower than predictions of current evolutionary models of massive stars. An imprecise mass-loss and distance were responsible for the much greater discrepancy in neon content identified by Barlow et al. Our sample of WC5--8 stars span a narrow range in T* (=55--71kK), with no trend towards higher temperature at earlier spectral type, supporting earlier results for a larger sample by Koesterke & Hamann (1995). Stellar luminosities range from 100,000 to 500,000 Lo, while 10^-5.1 < Mdot/(Mo/yr) < 10^-4.5, adopting clumped winds, in which volume filling factors are 10%. In all cases, wind performance numbers are less than 10, significantly lower than recent estimates. Carbon abundances span 0.08 < C/He < 0.25 by number, while oxygen abundances remain poorly constrained.

astro-ph↗