Previous
abstract 
Next
abstract
It has been recognized for a decade that stellar models which account for the microscopic diffusion of lithium are not compatible with the nearly uniform lithium abundance, as a function of temperature, found in metal poor halo dwarfs (the Spite plateau). The recent observations of Thorburn (1994) extend the Spite plateau by about 200 degrees (to higher temperatures) and thus significantly exacerbate the problem. Herein it is shown that microscopic diffusion can be compatible with the lithium observations if mild main sequence mass loss occurs throughout the evolution of these halo dwarfs. Mass loss rates of $\sim -1\times 10^{-12} M_{\scriptscriptstyle\odot} \ {\rm yr}^{-1}$ are required. These models suggest that halo stars have depleted their initial lithium uniformly by a factor of $\sim 1.6$ or 0.2 dex. Since the diffusion of helium and metals are similarly affected, these models should also provide agreement with globular cluster color-magnitude diagrams superior to that of diffusion models without mass loss.
Besides providing good agreement with dwarf lithium data of Thorburn (1994, ApJ, 421, 318), models with diffusion and mass loss provide excellent agreement with the population II sub-giant lithium data collected by Pilachowski, Sneden, \& Booth (1993, ApJ, 407, 699). Also, such models naturally explain why the dispersion in lithium abundance is greater for the sub-giants relative to the dwarfs. It is will be shown that observations of hot sub-giants can potentially distinguish between the inhibition of diffusion by mass loss versus its inhibition by turbulent or rotation mixing.
S. Vauclair \& C. Charbonnel (1994, preprint) have independently considered such mass loss and have computed appropriate main-sequence models. They obtained nearly identical results with regard to the impact of similar mass loss rates on diffusion in extreme population II dwarfs. A paper on my work appears as Los Alamos report LA-UR-94-2961. That paper was submitted to ApJ letters but may end up elsewhere (possibly ApJ).
