Extremely metal-poor stars .3. The lithium-depleted main-sequence turnoff dwarfs

We present abundances of 14 elements for the metal-poor, near-main-sequence turnoff stars G66-30, G139-8, and G186-26, which are well known to possess less than 1/5-1/10 the value of Li/H observed in field halo Spite Plateau dwarfs and are thought by many to have been produced in the currently accepted standard big bang cosmology. The stars have [Fe/H] = -1.53, -2.24, and -2.68, respectively, and for these values their collective abundance patterns are not abnormal. That is to say, there is no common abundance abnormality that one might associate with their Li deficiencies. In G186-26, we find an overabundance of the heavy neutron-capture elements that increases with atomic mass. We measure [Ba/Fe] = +0.35, whereas most stars of this metallicity have [Ba/Fe] similar to -0.5, together with smaller enhancements for Sr and possibly Y. Such supersolar, neutron-capture, element abundances, however, are exhibited by similar to 25% of stars having [Fe/H] similar to -2.7. No such enhancement is found in G66-30 or G139-8, and the simplest explanation for the data of the three stars is that there is no connection between Li depletion and the abundance patterns of the heavy neutron-capture elements. G66-30 and G186-26 appear to be multiple systems, and we discuss the possible role of binarity in producing the observed Li depletion. We concur with J. A. Thorburn that this group of objects is not the progeny of blue stragglers, and we discuss the alternative that some of the Li-depleted stars may be the Population II counterparts of the disk population subgiant CH, barium dwarf, and F str lambda 4077 stars. As suggested by others, the most likely explanation for the abundance anomalies in these objects, which includes universal Li deficiency, involves mass transfer across a binary during the asymptotic giant branch evolutionary phase of the erstwhile primary of the system. We note that while such mass transfer might be expected to produce Li depletion variations in C, N, and the heavy neutron-capture elements are possible but not all necessary. Neither G66-30 nor G139-8 shows enhancement of C or the neutron-capture elements: and no information is available for N. Further work is necessary to settle the issue. If mass transfer is responsible for Li depletions, such stars justifiably could be excluded from investigations of the primordial Li abundance, In the absence of evidence for such mass transfer, however, the possibility remains that some of these objects, if not all, are extreme examples of a process that has affected all Plateau stars.