The intrinsic dispersion of elemental abundance ratios in nearby metal-poor halo stars

TL;DR

This study measures elemental abundance ratios in 25 nearby metal-poor halo stars, revealing very low intrinsic scatter and correlations among certain elements, which inform models of early Galactic chemical evolution.

Contribution

It provides precise measurements of elemental abundance dispersions and correlations in metal-poor stars, improving understanding of nucleosynthesis and chemical evolution.

Findings

Intrinsic dispersion of [X/Fe] is very small, less than 0.07 dex for many elements.

Strong correlations found among residuals of alpha-elements and between Y and Zr.

Differential atomic diffusion explains some residual correlations, but stochastic supernova effects are also important.

Abstract

Differential abundances of C, O, Mg, Al, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Zn, Y, and Zr were determined from high signal-to-noise VLT/UVES spectra for 25 turnoff stars with -2.4 < [Fe/H] < -1.3. Effective temperatures were obtained from profiles of the H_beta line and surface gravities via Gaia parallaxes. The analysis of the spectra were based on 1D model atmospheres assuming LTE, but 3D non-LTE corrections were applied for several elements. The dispersion in linear fits to the [X/Fe]-[Fe/H] relations is around a factor of two smaller than found in previous studies. After corrections for measurement errors, the 1-sigma intrinsic dispersion of [X/Fe] at a given metallicity is 0.09 dex for Y and Zr, 0.05-0.07 dex for C, O, and Al, 0.03-0.05 dex for Mg, Ca, Sc, Ti, V, Mn, and Zn, and <0.03 dex for Cr, Co, and Ni. Strong correlations between the residuals in the [X/Fe]-[Fe/H] fits are found for the alpha-capture elements Mg, Al, Ca, Sc, and Ti and between the residuals for Y and Zr. Correlations of the residuals in the [X/Fe]-[Fe/H] fits with effective temperature can be explained as due to differential atomic diffusion between elements, but its contribution to the scatter of [X/Fe] is of minor importance. Probably, both stochastic effects in sampling the IMF of CCSNe and differences in the TypeIa to CCSNe enrichment ratio between star-forming regions need to be considered in order to explain the intrinsic dispersion of [X/Fe].