Vega

Abstract: Photometric observations at optical and near-infrared wavelengths are presented for members of a new sample of candidate Vega-like systems, or main sequence stars with excess infrared emission due to circumstellar dust. The observations are combined with IRAS fluxes to define the spectral energy distributions of the sources. Most of the sources show only photospheric emission at near-IR wavelengths, indicating a lack of hot (,1000 K) dust. Mid-infrared spectra are presented for four sources from the sample. One of them, HD 150193, shows strong silicate emission, while another, HD 176363, was not detected. The spectra of two stars from our previous sample of Vega-like sources both show UIR-band emission, attributed to hydrocarbon materials. Detailed comparisons of the optical and IRAS positions suggest that in some cases the IRAS source is not physically associated with the visible star. Alternative associations are suggested for several of these sources. Fractional excess luminosities are derived from the observed spectral energy distributions. The values found are comparable to those measured previously for other Vega-like sources. Key words: circumstellar matter ± planetary systems ± infrared: stars. 1 I N T R O D U C T I O N Vega-excess, or Vega-like systems are main sequence stars that exhibit infrared emission above expected photospheric levels. The best known Vega-excess stars are a Lyr itself and b Pic. The excess flux is ascribed to thermal emission from circumstellar dust grains orbiting in a disc or ring structure (see e.g. Aumann et al. 1984). Imaging of the dust emission at infrared and submillimetre wavelengths (e.g. Holland et al. 1998; Koerner et al. 1998) and coronagraphic imaging of scattered light (e.g. Smith & Terrile 1984; Schneider et al. 1999) have confirmed that the dust is indeed distributed in discs. Many searches of the IRAS catalogues have found other candidate Vega-like stars (e.g. Aumann 1985; Sadakane & Nishida 1986; Walker & Wolstencroft 1988; Stencel & Backman 1991). The Infrared Space Obsevatory (ISO) has also been used to search for new Vega-likes. One notable result was the discovery of dust emission from r Cnc (Dominik et al. 1998), a G8V star which is known to host at least one planet (Butler et al. 1997). Comprehensive reviews of the Vega-excess phenomenon and its relation to the formation of planets can be found in Backman & Paresce (1993), Ferlet & Vidal-Madjar (1994) and Lagrange, Backman & Artymowicz (2000). 1.1 The new catalogue Mannings & Barlow (1998; henceforth MB) have recently published the results of a new search for candidate Vega-like systems. The search made use of the IRAS Faint Source Survey Catalogue (FSC; see Moshir et al. 1992 for details). The FSC has a sensitivity limit approximately one magnitude fainter than the Point Source Catalog (PSC), achieved by co-adding the individual detector data before extracting sources. The increased sensitivity is gained at the expense of a slight reduction in the reliability of the detections (94 per cent, compared with 99.997 per cent for the PSC). The FSC was searched for main-sequence stars by extracting sources that were positionally associated with luminosity class V stars from the published volumes of the Michigan Catalog of TwoDimensional Spectral Types for the HD Stars (Houk & Cowley 1975; Houk 1978; Houk 1982; Houk & Smith-Moore 1988). Volumes 1±4 of the Michigan catalogue give spectral types and luminosity classes for some 130397 HD stars south of a declination d ˆ 2128. Cross-correlating these catalogues gave a total of 294 luminosity class V stars that have good detections in the IRAS bands 1± 3, 1 and 2, or 2 and 3 (bands 1, 2, 3 and 4 have wavelengths of 12, 25, 60 and 100mm respectively). Of these, 131 stars were found to show significant IR excess emission. See MB for details of the selection process. q 2000 RAS w E-mail: rjs@star.ucl.ac.uk 74 R. Sylvester and V. Mannings MB searched for excess IR emission on the basis of ratios of the IRAS fluxes, reasoning that a fall-off of flux with wavelength that was shallower than that expected from a stellar photosphere implies excess emission. They therefore did not need to determine the expected (i.e. photospheric) fluxes in the IRAS bands. While this approach provides a robust method for identifying sources with excess emission, detailed analysis of the emitting material requires knowledge of the photospheric contribution to the IR fluxes. We have therefore made optical and near-IR photometric observations of a number of the MB sources, in order to characterize their spectral energy distributions (SEDs) at wavelengths where thermal dust emission will be negligible, which allows the photospheric contribution to the SED to be calculated for all wavelengths (typically by using a model atmosphere) and to find the wavelength at which the onset of excess emission occurs. The latter is important for deriving the total excess luminosity, and the maximum temperature of the emitting dust. 2 O B S E RVAT I O N S 26 stars from the MB sample were selected for observing on the basis of their RA and such photometry as was available in the literature. These target stars are listed in Table 1. All the new photometric observations presented here were made in 1995 September, using the facilities of the South African Astronomical Observatory. Spectra in the 10mm band were obtained at UKIRT in 1996 September and 1998 July. 2.1 Optical photometry Photometry was obtained in both the broad-band UBV(RI)C system and in the intermediate-band StroÈmgren uvby system. Both sets of observations were made using the Modular Photometer on the 0.5-m reflector at the SAAO Sutherland site. See Kilkenny et al. (1998) and Kilkenny & Laing (1992) for details of the use of this instrument, the adopted standard star systems and the data reduction procedures for UBV(RI)C and uvby photometry respectively. An aperture of 25 arcsec was used, and sky background measurements were made for each star that was observed. The data were reduced by SAAO staff, using the photometric reduction procedures described in the latter two references. 2.2 Near-infrared photometry Near-IR (JHKL) observations were made with the MkII Infrared Photometer at the Cassegrain focus of the 0.75-m reflector at Sutherland. The MkII photometer uses an InSb detector cooled by pumped liquid nitrogen, and is of similar design to the MkI instrument described by Glass (1973). Background subtraction is achieved using a sector-wheel chopper in the instrument, with a chopping frequency of 12.5 Hz. The photometer integration time was 10 seconds at a single pointing. The telescope was nodded in declination every 20 s to allow for gradients in the sky emission. An observing `module' consists of two nodded pairs of integrations (on-off-off-on), and therefore lasts 40 seconds. For each target, sufficient modules were taken to attain an adequate signalto-noise ratio in the co-added result. 2.3 Hipparcos/Tycho photometry As well as a wealth of parallax and astrometric data, the Hipparcos mission (ESA 1997) produced a great deal of useful photometric data, for the pre-selected stars in the Hipparcos Input Catalogue, and also for more than a million stars brighter than V , 11:5 (forming the Tycho Catalogue). Stars in the Hipparcos catalogue have photometry in the B, V and I bands, derived from satellite and selected ground-based measurements, while the Tycho catalogue gives B and V magnitudes. Data from the Hipparcos and Tycho catalogues were used to supplement the new ground-based photometry and to check our results. 2.4 Mid-IR spectroscopy Mid-IR spectra were taken on 1998 July 1 of four sources from the MB catalogue: HD 109085, 123356, 150193 and 176363. The data were obtained at UKIRT, using the common-user helium-cooled spectrometer CGS3 in its low-resolution …R < 55† 10mm mode. A detailed description of CGS3 can be found in Cohen & Davies (1995). The spectra were flux calibrated by dividing the spectrum of the target by that of a standard star, then multiplying by a model of the mid-IR spectrum of the standard. a Boo was used to calibrate the HD 109085 spectrum; the model for this standard was the absolutely-calibrated spectrum of Cohen et al. (1995). HD 123356 and HD 150193 were calibrated using s Lib, assumed to radiate as a 3640 K blackbody with a 10.0-mm flux of 188.6 Jy, while HD 176363 was calibrated using h Sgr, treated as a 3600 K blackbody with a 10.0-mm flux of 208.7 Jy. Many of the stars in the MB list are too far south to be observed from UKIRT, so we have not attempted a comprehensive CGS3 survey. Similarly, we do not have SAAO photometry for two of the stars observed with CGS3 due to the different constraints on the two observing runs. The four MB stars observed with CGS3 were selected on criteria including their declination, IRAS 12-mm flux and compatibility with the scheduling of other targets on the observing night. q 2000 RAS, MNRAS 313, 73±86 Table 1. Log of the photometric observations. HD SAO HR Other Sp. T UBV uvby nIR 7151 215402 F6V p p p 1080

Abstract: Context. Among optical stellar interferometers, the CHARA Array located at Mt Wilson in California offers the potential of very long baselines (up to 330 m) and the prospect of coupling multiple beam combiners. This paper presents the principle and the measured performance of VEGA, Visible spEctroGraph and polArimeter installed in September 2007 at the coherent focus of the array.Aims. With 0.3 ms of arc of spatial resolution and up to of spectral resolution, VEGA intends to measure fundamental parameters of stars, to study stellar activities and to image and analyze circumstellar environments. We describe the observing modes that have been implemented for this spectro-polarimeter and show actual performances measured on the sky during the first observing runs.Methods. The astrophysical programs are described in relation to the observing modes of the instrument, the presentation of the spectrograph and of the interface table is shown and finally the data is presented. We discuss the perspectives of further development in the framework of the CHARA Array.Results. We show that VEGA/CHARA is fully operational. The current limiting magnitude is nearly 7 but the results depend on the observing conditions (seeing, spectral resolution, etc.). We have validated the stability of the instrumental visibility at the level of 1 to 2% over half an hour and of the instrumental polarization for various declinations. Some examples of squared visibility and differential visibility are presented.Conclusions. The spectro-polarimeter VEGA has been installed and successfully tested on CHARA. It will permit stellar physics studies at unprecedented spectral and spatial resolutions.

Abstract: We have obtained high-precision interferometric measurements of Vega with the CHARA Array and FLUOR beam combiner in the K' band at projected baselines between 103m and 273m. The measured visibility amplitudes beyond the first lobe are significantly weaker than expected for a slowly rotating star characterized by a single effective temperature and surface gravity. Our measurements, when compared to synthetic visibilities and synthetic spectrophotometry from a Roche-von Zeipel gravity-darkened model atmosphere, provide strong evidence for the model of Vega as a rapidly rotating star viewed very nearly pole-on. Our best fitting model indicates that Vega is rotating at ~91% of its angular break-up rate with an equatorial velocity of 275 km/s. Together with the measured vsin(i), this velocity yields an inclination for the rotation axis of 5 degrees. For this model the pole-to-equator effective temperature difference is 2250 K, a value much larger than previously derived from spectral line analyses. The derived equatorial T_eff of 7900 K indicates Vega's equatorial atmosphere may be convective and provides a possible explanation for the discrepancy. The model has a luminosity of ~37 Lsun, a value 35% lower than Vega's apparent luminosity based on its bolometric flux and parallax, assuming a slowly rotating star. The model luminosity is consistent with the mean absolute magnitude of A0V stars. Our model predicts the spectral energy distribution of Vega as viewed from its equatorial plane; a model which may be employed in radiative models for the surrounding debris disk.

Abstract: We have obtained high-precision interferometric measurements of Vega with the CHARA Array and FLUOR beam combiner in the K’ band at projected baselines between 103m and 273m. The measured visibility amplitudes beyond the first lobe are significantly weaker than expected for a slowly rotating star characterized by a single effective temperature and surface gravity. Our measurements, when compared to synthetic visibilities and synthetic spectrophotometry from a Roche-von Zeipel gravitydarkened model atmosphere, provide strong evidence for the model of Vega as a rapidly rotating star viewed very nearly pole-on. Our model of Vega’s projected surface consists of two-dimensional intensity maps constructed from a library of model atmospheres which follow pole-to-equator gradients of effective temperature and surface gravity over the rotationally distorted stellar surface. Our best fitting model, in good agreement with both our interferometric data and archival spectrophotometric data, indicates that Vega is rotating at ∼91% of its angular break-up rate with an equatorial velocity of 275 km s −1 . Together with the measured v sin i, this velocity yields an inclination for the rotation axis of 5 ◦ . For this model the pole-to-equator effective temperature difference is 2250 K, a value much larger than previously derived from spectral line analyses. A polar effective temperature of 10150 K is derived from a fit to ultraviolet and optical spectrophotometry. The synthetic and observed spectral energy distributions are in reasonable agreement longward of 140 nm where they agree to 5% or better. Shortward of 140 nm, the model is up to 10 times brighter than observed. The far-UV flux discrepancy suggests a breakdown of von Zeipel’s Teff ∝ g 1/4 relation. The derived equatorial Teff of 7900 K indicates Vega’s equatorial atmosphere may be convective and provides a possible explanation for the discrepancy. The model has a luminosity of ∼37 L⊙, a value 35% lower than Vega’s apparent luminosity based on its bolometric flux and parallax, assuming a slowly rotating star. The model luminosity is consistent with the mean absolute magnitude of A0V stars from the W(H) − MV calibration. Our model predicts the spectral energy distribution of Vega as viewed from its equatorial plane; a model which may be employed in radiative models for the surrounding debris disk. Subject headings: methods: numerical — stars: atmospheres — stars: fundamental parameters (radii, temperature) — stars: rotation — stars:individual (Vega) — techniques: interferometric