The intermediate-band solar cell (IBSC) concept has been recently proposed to enhance the current gain from the solar spectrum whilst maintaining a large open-circuit voltage. Its main idea is to introduce a partially occupied intermediate band (IB) between the valence band (VB) and conduction band (CB) of the semiconductor absorber, thereby increasing the photocurrent by the additional $\text{VB}\to \text{IB}$ and $\text{IB}\to \text{CB}$ absorptions. The confined electron levels of self-assembled quantum dots (QDs) were proposed as potential candidates for the implementation of such an IB. Here we report experimental and theoretical investigations on ${\text{In}}_y{\text{Ga}}_{1-y}\text{As}$ dots in a ${\text{GaAs}}_{1-x}{\text{P}}_x$ matrix, examining its suitability for acting as IBSCs. The system has the advantage of allowing strain symmetrization within the structure, thus enabling the growth of a large number of defect-free QD layers, despite the significant size mismatch between the dot material and the surrounding matrix. We examine the various conditions related to the optimum functionality of the IBSC, in particular those connected to the optical and electronic properties of the system. We find that the intensity of absorption between QD-confined electron states and host CB is weak because of their localized-to-delocalized character. Regarding the position of the IB within the matrix band gap, we find that, whereas strain symmetrization can indeed permit growth of multiple dot layers, the current repertoire of ${\text{GaAs}}_{1-x}{\text{P}}_x$ barrier materials, as well as ${\text{In}}_y{\text{Ga}}_{1-y}\text{As}$ dot materials, does not satisfy the ideal energetic locations for the IB. We conclude that other QD systems must be considered for QD-IBSC implementations.

• Received 20 August 2008

DOI:https://doi.org/10.1103/PhysRevB.78.205321