2006MNRAS.372L..48J

The identification of non-radial g-mode oscillations as the cause of variability in cool subdwarf B stars (PG1716 variables) has been frustrated by a 5000-K discrepancy between the observed and theoretical blue-edge of the instability domain. A major component in the solution to this problem has been identified by (i) using updated Opacity Project (OP) instead of OPAL opacities and (ii) considering an enhancement of nickel, in addition to that of iron, in the driving zone. The reason for this success is that, in OP, the `Fe-bump' contributions from iron and nickel occur at higher temperatures than in OPAL. As well as pointing to a solution of an important problem in stellar pulsation theory, this result provides a critical test for stellar opacities and the atomic physics used to compute them.