HR 5110: An Algol system with RS CVn characteristics

Abstract

HR 5110 (HD 118216=BH CVn) is a close binary system (P=2~6) which is viewed nearly pole-on (i=13~). A comparison of the characteristics of Algol and RS CVn systems to those of HR 5110 shows that HR 5110 can also be considered an Algol system. Because the primary star is relatively cool (F2 IV) and there is no apparent emission from an accretion disk, we are able to detect in IUE spectra the emission of an active chromosphere and transition region of the cooler secondary. HR 5110 is the only known Algol system for which the properties of the secondary star can be studied in detail. For a more complete discussion see Little-Marenin et al. (1986). An interpretation of HR 5110 depends critically on knowin~ the spectral type and luminosity class of the two stars in the system. To accomplish this we obtalne~ near-infrared JHKLM photometry of HR 5110 on the 1.3-m telescope of the Kitt Peak National Observatory in 1980 February and June and also 1981 May. The best f~t to the observed colors of HR 5110 for different types of primaries (F2 V to GO V) and Secondaries (G5 IV to K1 IV) is a F2 IV primary similar to o Boo and a KO IV secondary (Vs-Vp=I~88) for which spots cover 35-40% of the surface. The determination that the KO IV star is spotted comes from the infrared excess (from J to M'), which is fairly well matched with a blackbody curve of 3100 K and therefore is too hot to be produced by dust and has the wrong wavelength dependence to be produced by free-free emission. Our spectral synthesis differs from that of Shore and Adelman (1984), but is in agreement with the statistics found for Algol systems by Giuriein, e t al. (1983); every F primary has a K secondary. We obtain from the Barnes-Evans relationship (Barnes and Evans 1976) an angular diameter of 0.47 milliarcsee for the spotted K0 IV star, corresponding to a radius of 1.9-2.7 R 8 for a parallax of 0~0.27 ~ ~ ~ 0~0.19. We conclude that the secondary fills its Roche lobe radius of 2.6 RS, as originally suggested by Conti (1967). We obtained IUE spectra during the latter (plateau) phases of the radio flares of 1979 May 29-31 and 1981 April 6, 56 hours after the first major event (Simon et al. 1980) and 26 hours after the second. We also observed HR 5110 in 1980 February and in 1983 August. The IUE archives yielded two additional spectra presumably taken when the system was quiescent. Unless an unobserved flare occurred on 1981 February, the IDIE data appear to show correlations with phase with maximum flux