HIGH ENERGY EMISSION FROM B STARS AND ITS RELATIONSHIP TO STELLAR WINDS

Abstract

ABSTRACT In this work I present results of detailed studies of the high energy emission of two B stars using the EUVE, ROSAT, and ASCA telescopes. I also present the results of a ROSAT survey of 27 near-main-sequence B stars. These observations and related efforts at modeling the data are used to address several outstanding questions about B stars, their high energy emission, and their stellar winds. These questions involve the applicability of standard radiation-driven wind theory to B stars, the ability of the line-force instability mechanism and other wind-shock mechanisms to account for the EUV and X-ray emission from B stars, and the need to invoke non-wind emission mechanisms for anomalous B stars and those B stars with spectral type later than about B2. The first part of this study involves the simultaneous analysis of an EUVE emission line spectrum and the ROSAT PSPC data of the normal B2 II star Epsilon CMa. I find that a continuous temperature distribution in the emitting plasma as well as attenuation by the overlying stellar wind is needed to explain both data sets. The attenuation is consistent with the predictions of the wind instability model, while the temperature distribution and emission measure may require some slight modifications in the applications of the theories of the line-force instability and radiation-driven winds, respectively, to Epsilon CMa. In order to analyze these data sets I performed calculations of the ionization structure of the stellar wind which, for the first time, took into account the influence of X-rays on the photoionization rates. I also performed numerical simulations of the wind of Epsilon CMa to explore the applicability of the line-force instability mechanism. The next part of the study is the ROSAT survey in which I detected 20 of 27 near-main-sequence B stars. This result is in direct contradiction to earlier surveys in which only five to ten percent of stars later than B1 were detected. I find that the X-ray luminosities fall rapidly beyond B1, but they do not disappear. It is nearly impossible for these observations of mid- and late-B stars to be understood within the context of nearly any reasonable wind-shock model. On the other hand, the early-B stars in this survey have temperature distributions and wind filling factors similar to those found for Epsilon CMa. The degree of wind attenuation in these stars is also significant. Because the spectral models I employ account for wind attenuation I am able to derive both the emergent and the generated X-ray luminosities. One implication of these results, and also of the model fits to Epsilon CMa is that the well known LX/LBol ~10-7 law is the result of wind attenuation more than it is the result of anything inherent to the emission process. Finally, I used moderate resolution ASCA data to study the high energy processes on the unusual B0 V star Tau Sco. This star is anomalous in several ways, including extreme youth, which may affect its X-ray properties. It has a high X-ray luminosity (LX/LBol ~10-6) and a very hard spectrum. The ASCA data reveal emission line complexes from very highly ionized iron, magnesium, silicon, and sulfur which help constrain the plasma temperature distribution. I show that there must be a significant quantity of gas with temperatures in excess of 28 million degrees, which is far too hot to be explained by line-force instability wind shocks. It seems that coronal mechanisms may need to be invoked to explain the high energy properties of some mid- and late-B stars as well as a small number of anomalous earlier-type OB stars. The inclusion of Tau Sco in this category indicates that youth may play an important role in this non-wind high energy mechanism, whatever it may turn out to be. Normal early-B stars, however, do not strain the current wind-shock theories too badly, as long as the mass loss rates have been underestimated by a factor of two or three. New calculations of the line force in these winds, taking the effects of X-ray photoionization into account, seems warranted at this time.