The Fundamental Physical Properties of Wolf-Rayet Stars

Massive stars are the cosmic engines that power the far-infrared luminosities of distant galaxies, and dominate the ionization of nearby HII regions. They are the primary source of carbon and oxygen in the Universe, and their core collapses manufacture all of the elements heavier than Fe. The re-ionization of the early Universe was thanks to Population III massive stars, and the super-massive black holes we find in the cores of galaxies today were seeded as a result of the black holes that formed from the first generations of massive stars. Understanding massive star evolution is the key to unlocking many astrophysical problems. The largest uncertainty in massive star evolution is the question of how Wolf-Rayet (WR) stars form. Our proposal will determine the fundamental physical properties of WRs using four archival NASA data sets for a critical comparison with present day evolution models. It is generally assumed that massive stars spend most of their post-main-sequence lives WRs. For decades we have believed that WRs form as a result of stellar winds stripping off the H-rich outer layers of a star, leaving behind a bare stellar core. In this picture, WRs are a normal stage in the evolution of the most massive stars. Recently, this scenario has been called into question. Stellar wind mass- loss rates are now known to be significantly lower than previously thought, although whether this is a factor of 3 or 10 remains unclear. If the latter is correct, then this poses a serious problem for the formation of WRs. This has created a paradigm shift, with increased importance attached to the role of binary evolution, with Roche-lobe overflow performing the stripping. Attempts to distinguish which scenario is more prevalent is complicated by the possibility of past mergers; i.e., just because a WR is not a binary today does not prove it was not one in the past. We will tackle this question from a fresh perspective, determining reliable fundamental physical properties of WRs and seeing whether they better match the single or binary star evolutionary models. If they agree with the single-star models, that is compelling evidence that WRs are a normal part of the evolution of massive stars. If they disagree, perhaps either binary evolution plays an important role in the formation of WRs or the single star models could be improved. For instance, we know that the mass-loss rates during the Luminous Blue Variable and red supergiant phases are poorly constrained by observations. If higher mass loss rates during these phases were included, could we account for all of the WR physical properties (including chemical abundances) that we find? Either result will help us learn more about the origin of WRs while also testing and helping improve the evolutionary models. For this test to be meaningful, we must have accurate measurements of the fundamental physical properties of WRs (such as effective temperatures, bolometric luminosities, and chemical abundances), as well as having a good understanding of the uncertainties on these quantities. To achieve this, we have a selected a statistically large sample of 27 WRs in the Small and Large Magellanic Clouds which possess excellent UV spectra in the MAST IUE archive. This wavelength is crucial, as it contains key diagnostic resonance lines, such as CIV 1550. To these, we add our own high quality Magellan optical (and, when needed, near-IR) spectrophotometry. Forty percent of our sample has also been observed in the far-UV with FUSE, providing additional diagnostics. Finally, we will incorporate NASA 2MASS and Spitzer IPAC photometry, which extend the spectral energy distribution into the IR. We will model each of these combined data sets using CMFGEN, a stellar atmosphere code that includes the many complications needed to model the spectra of these stars. The use of this combined data set achieves what one could not hope to do from any one of them, consistent with the aims of the ADAP.