Abundance analysis of selected cepheids and the galactic distribution of metallicity

Abstract

We have determined the atmospheric abundances of selected Cepheids in order to study the large-scale chemical inhomogeneities across the galactic disk. The classical Cepheids were selected as probes to study the variation of metallicity in the galactic disk, because of their high intrinsic luminosity, small age and the existence of period-luminosity and period-age relationships. High dispersion spectra of programme stars WZ Sgr, X Sgr, ς Gem, T Mon and S V Mon were obtained using the 102-cm reflector of Kavalur Observatory. The atmospheric abundances were determined by theoretically synthesizing the selected portions of the stellar spectrum and comparing with the observed spectra. In order to compute the theoretical spectrum, the formal solution of the equation of radiative transfer was numerically evaluated with the simplifying assumptions of local thermodynamical equilibrium, plane-parallel geometry and hydrostatic equilibrium. These assumptions are reasonably good for the metallic lines of F-G supergiants and hence the observations were confined to the phases where Cepheids behave like nonvariable F-G supergiants. The atmospheric abundances of iron-peak elements, Fe, Cr, Ti, Ca and heavier s-process elements Y, Ba, La, Ce, Sm were obtained by synthesizing a selected spectral region in the range 4330 A — 4650 A. We derive a radial abundance gradient for iron $$\frac{{d(Fe/H)}}{{dr_{gc} }} = - 0.056 \pm 0.08$$ for the region of galactic disk between 6.7 and 10.9 kpc from the galactic centre (assuming rgc = 8.5 kpc for the Sun). This value agrees with the one obtained from the general sample of Cepheids for which spectroscopic abundances are available, and also with the existing photometric determinations, but is shallower than the one derived by Luck (1982). Abundances of the elements derived in the present investigation do not show any significant correlation with atomic number. Also the abundance ratio of s-process elements does not show any correlation with Fe. This lack of correlation for disk population stars shows the inadequacy of simple models of galactic chemical evolution and favours the infall models. Alternately, the evolution of [s/Fe] may be determined by the ratio of intermediate-mass stars (which contribute s-process nuclei) to high-mass stars (which contribute Fe peak nuclei). Thus the different behaviour of halo and disk population may indicate a difference in the mass spectrum of star formation.