Heavy metal opacity and line blanketing in hot DA white dwarf stars

This thesis concerns atomic data, and it’s impact upon white dwarf model atmosphere calculations, and consequentially, the measurements made using such models. The thesis begins with a brief introduction to the history and properties of white dwarfs, the theory of radiative transfer, and the uses of white dwarfs in astronomy. A detailed spectroscopic survey of WD0501+524 (G191-B2B) is presented. 976 absorption features were detected, 947 of which have been successfully identified. ∼ 60% of the identified features were found to pertain to Fe and Ni iv-vi transitions. The potential consequences of using a limited atomic data set in model atmospheres are discussed, and other possible identifications to the unknown absorption features are queried. The Kurucz (1992) (Ku92) atomic database (containing ∼ 9, 000, 000 transitions) is supplemented by photoionisation (PI) cross section data from the Opacity Project for use in stellar atmosphere calculations. The more recent Kurucz (2011) (Ku11) database (containing ∼ 160, 000, 000 transitions) is not accompanied by PI cross section data. Calculations performed to create this data are described, and their effects on model atmosphere calculations are discussed. The Lyman/Balmer line problem, a discrepancy between the measured effective temperatures (Teff) from the Lyman/Balmer line series is considered. Teff and surface gravity (log g) measurements are made of 24 DA white dwarfs using model atmosphere grids utilising two atmospheric compositions, two Stark broadening tables, and the Ku92 and Ku11 atomic data sets. It is shown that the average opacity contributed by all metal species in the atmosphere drives the discrepancy between the measured Lyman/Balmer Teff. Analysis of Extreme Ultraviolet Explorer (EUVE) data for seven metal rich white dwarf stars is presented. Four model atmosphere grids were calculated using two atmospheric compositions (Preval et al. 2013; Barstow et al. 2003) and the Ku92 and Ku11 atomic datasets. Improved fits for wavelengths shortward of 230°A are obtained for all stars except WD0501+524, where significant discrepancies remain.