A comparative analysis of the laboratory and theoretical transition probabilities of the Fe-peak elements for a new release of VALD

Abstract

We carried out a comparative analysis of the recent atomic data for iron-peak elements, mainly Ti, Cr and Fe, for a new release of the Vienna Atomic Line Database (VALD3). New data were compared with those available in VALD2 and were checked using high-resolution, high signal-to-noise spectra of sharp-lined chemically normal stars including the Sun, and the zero-rotation extremely Cr- and Fe-rich chemically peculiar star HD 133792. The observed spectrum of the latter star allowed for comparison with transition probability calculations based on the orthogonal operator technique with the Cowan code for Cr II and Fe II lines for lower level energies between 2 eV and 11 eV in the wavelength region 3100 to 9000 Å.In general, the agreement between the new experimental transition probabilities and those currently available in VALD2 is fairly good, which helps to validate the stellar abundance data derived with the VALD2 atomic parameters. We also found that, for a few important Ti II and Fe II lines in the visible spectral region, new transition probabilities are not consistent within their quoted accuracy.In a series of recent works on experimental f-values for Fe II it was shown that calculations based on the orthogonal operator technique agree better with the experimental data than the Cowan code calculations and, hence, should have preference for stellar spectroscopy. Our analysis of the Ap star HD 133792 spectrum clearly demonstrates that there are quite a number of high-excitation Cr II and Fe II lines which are fitted reasonably well when using the transition probabilities calculated with the Cowan code. As a rule these lines have their upper energy levels classified differently in both methods of calculations.