Charge-Transfer Excitations: A Challenge for Time-Dependent Density Functional Theory That Has Been Met

Correctly predicting charge-transfer states in complex molecules and molecular materials has been a major challenge for first-principles theory. With the computational cost of accurate ab initio wave-function methods often being too high for exploring systems of technological relevance, one has to resort to time-dependent density functional theory (TDDFT). However, for a long time the incorrect description of charge-transfer excitations has been considered one of the hallmark failures of TDDFT. As charge-transfer states play an important role in organic solar cells, and generally in many new materials that are aimed at converting sunlight into other forms of energy, the contribution of first principles theory to the field of energy relevant materials was seriously limited. However, in the past years this limitation has been overcome. A new class of nonempirical density functionals can predict charge-transfer excitations in molecules reliably. The approach by now has been extended to describe molecular solids and solvated systems. The predictive power of TDDFT has thus greatly increased, and computational studies can be expected to give true guidance in material design.