Abstract. Defects in semiconductors and oxides have been the subject of some of the most sophisticated approaches to modelling and simulation. The powerful, widely used methods can give the impression that all technologically important materials problems can be addressed reliably. But is this so? This paper looks at some of the gaps in first principles theories and at the situations that still warrant attention. Excited states, non-equilibrium systems and non-adiabatic transitions, the correct handling of different length and time scales and the prediction of characteristically quantum behaviour all present challenges. Whilst the emphasis is on semiconductor and oxide systems, the wider context of materials science points to further issues that should not be overlooked.

Print publication: Issue 8 (December 2009)
Received 7 May 2009, in final form 5 July 2009
Published 23 November 2009