Abundances of the elements He to Ni in the atmosphere of Sirius A

Abstract

Context. The abundances of elements in the atmosphere of an individual star provide a very broad range of extremely important information about the formation and evolution history of that star. Specifically, during the main sequence phase, the abundance pattern observed can provide valuable constraints on a range of hydrodynamical processes occurring beneath the visible atmosphere. Aims. Among middle main sequence (A and B) stars, it is unusual to have abundance determinations for more than about a dozen of the 30 lightest chemical elements, largely because many elements do not have usefully strong spectral lines in the visible wavelength window usually used for abundance analysis. Exceptionally, for the bright, sharp-lined hot Am star Sirius A = HD 48915, high resolution, high signal-to-noise ratio observations of the spectrum are available from 1265 A to about 1 μ m. The aim of this project is to determine generally how useful the UV part of the spectrum is for “generic” abundance determination in A stars, and which abundances of light and iron peak elements can be determined or improved for Sirius A using this nearly unique data set. Methods. The UV and visible spectrum of Sirius A has been searched for useful spectral lines of all elements up to nickel (2 ≤ Z ≤ 28), as well as the heavy elements Sr, Y, Zr and Ba. Using the spectrum synthesis code ZEEMAN, abundances or upper limits are derived.Results. Abundances relative to H have been determined for a total of 20 elements, and upper limits are found for 4 others. Useful measurements or limits are possible from these data for all but three of the elements through Ni. In particular, it is found that the light element B is underabundant relative to the Sun by a factor of at least 30. Conclusions. Recent computations have led to predictions of the abundance pattern of almost all the elements having Z ≤ 28 expected in the atmosphere of Sirius as a result of competition between (1) gravitationally and radiatively driven diffusion and either (2) turbulent mixing or (3) homogeneous mass loss. Comparison of these predictions with observations has been hampered by the limited number of elements for which accurate abundance determinations are available. The work presented here increases the number of elements for which reasonably well-determined abundances are available, and establishes more clearly the uncertainty of the abundance values for most of the elements up to Ni. The new data provide useful tests of the theoretical predictions of the consequences of atomic diffusion in slowly rotating A stars, and in some cases contradict them. These data may also provide some fresh insight into the evolution of the Sirius system. It appears that the new abundance values may be qualitatively consistent with the possibility that significant mass transfer has occurred between the two components of the system during its evolution.