Binaries Danny Chris Carrie Ian Thomas Augusteijn Ian

Binaries Danny, Chris, Carrie, Ian Thomas Augusteijn, Ian Howarth

Why bother with binares? • Binary fraction is high so we need to understand them in order to understand populations • Binaries provide alternative evolutionary channels to single stars, perhaps more important channels at low Z (high z) • …and for understanding distribution of stars in HR diagram and distribution of vsini

NGC 346 -013: B 1 + O? , SB 2, Secondary is hotter, He. II present, and fainter in the optical. Large radial velocity variations, short time scale -> massive short period binary

NGC 346 -013: B 1 + ? , SB 2, Secondary is hotter (He. II present), and more massive.

Summary of dynamical information NGC 346 -013 system parameters: • Systemic velocity +156 km/s, i. e. the cluster velocity -> cluster member -> age of a few million years • P~4. 2 d, KB=250 km/s, KO=158 km/s, asini=33. 7 R סּ • MBsin 3 i=11. 4 M סּ MOsin 3 i=18. 1 M סּ • Effective Roche radii are Rl, B=10. 7 R סּ and Rl, O=13. 1 R סּ • (vsini)B=130 km/s -> if synchronised rotation/orbital periods RBsini = 10. 8 R סּ compare with Rl, B • If we adopt this radius then logg~3. 4 (spectroscopic logg=3. 5 ± 0. 2) Is the B star filling its Roche lobe? ?

Spectroscopic analysis of the B-star: Use red unblended wings of Balmer lines for logg, Si lines for temperature. Assume veiling factor for O-star (relative brightness) and iterate. Estimate that the ´O-star´ contributes 30 -40% of the flux in the B band. vsini determined from the Si III lines to be 130 km/s.

Abundances determined from the TLUSTY grid. Spectra were coadded in the B-star’s frame of reference.

NGC 346 -013: The B-star is nitrogen rich by about a factor of 10 #13 error SMC Teff Log g 25500 3. 50 1500 0. 2 - Vturb (km/s) [N/H] [O/H] [Mg/H] [Si/H] 6 7. 54 7. 81 6. 50 6. 63 3 0. 15 0. 18 0. 15 0. 34 6. 50 8. 05 6. 75 6. 80

What can we say about the O-star? • vsini = 320 km/s from He II 4686 • Difficult to see He. I lines from the Ostar, estimate Teff > 35 000 K • If relative flux f. O: f. B is 1: 2 this gives RO/RB ~? ? • The ‘O-star’ is not in co-rotation (it’s vsini implies R=26. 5 R )סּ • What is it?

Possibilities for this system n n The vsini is wrong and the system is a beta Lyr variable eclipsing binary like HD 163181 or V 443 Cygni Problem: most beta Lyr systems have similar spectral types n The B-star is the mass donor n The O-star has accreted a significant amount of mass and has been spun-up • Need the light-curve to constrain the radii. Proposal to WFI in early 2007. Submitted a FUSE proposal with Myron Smith to get UV spectra to better constrain the nature of the Ostar.

NGC 346 -029: B 0 V+? , v_sys = +188+/-1 km/s, semi-amplitude = 82+/-2 km/s, P=4. 47 d, [N/H]<6. 99; systemic velocity is high

NGC 346 -039: B 0. 7 V, vrad = +156. 5 +/- 3. 9 km, cluster member, [N/H]<6. 37 P~14. 6 d Semi-amplitude ~ 8 km/s Binary or pulsation?

NGC 346 -040: B 0. 2 V+? , vrad = +196. 6 +/- 44. 4 km/s, Binary with period ~ 100 d, semi-amplitude ~ 60 km/s

NGC 346 -040: B 0. 2 V+? , vrad = +181. 3 +/- 3. 6 km/s, probably binary, [N/H]<6. 88 (no period yet)

NGC 346 -116: B 1 V, vrad = +157. 0 +/- 2. 1 km/s, [N/H]=6. 93 Cluster member, and single.

Progress report • A lot of work on understanding systematics of FLAMES radial velocity measurements (paper in ESO VLT calibration workshop – Evans et al) • ESO proposals for WFI and FLAMES failed. . . too much competiton with ourselves. Need WFI (imaging) for short period systems and FLAMES for long period systems and supergiants. • FUSE satellite has failed…now considering COS and/or STIS for UV spectra of interesting systems post SM 4 (Aug 2008)