Submitted Abstract
Recombination processes inside a solar cell reduce the number of photogenerated charge carriers and thus directly reduce the electrical power output of the solar cell. Besides intrinsic recombination, which fundamentally occurs in all semiconductors, recombination processes mainly involve charge transition levels originating from electronic defects in the solar cell. Interfaces represent areas of reduced crystal symmetry and disturbed chemical bonds, and thus potentially contain a very high number of recombination-active defects. Dielectric surface passivation layers deposited onto the semiconductor surface allow a two-fold control over the interface recombination rate: altered chemical bonds at the interface reduce to total number of defects (“chemical passivation”), while fixed charges in the passivation layer repel one carrier type from the interface (“field-effect passivation”). We aim to exploit both of these passivation schemes to improve the efficiency of Cu(In,Ga)Se2 thin film solar cells by using thin aluminum oxide (Al2O3) passivation layers deposited by atomic layer deposition. We will study the energetic distribution of defect levels and their recombination dynamics, as well as the fixed charge density at the interface, by means of electrical and optical measurements. These results will be related to the chemical composition at the interface and details of the processing conditions to reveal the chemical nature of the dominant recombination defects and to establish a predictive model of interface recombination in this material system. These fundamental studies will be invaluable for future exploitation and optimization of surface passivation concepts in thin film photovoltaics.We will keep in mind that such compact and insulating passivation layers complicate the fabrication of electrical contacts to the semiconductor. Our approach will be to explore whether the passivation layer could be made thin enough to combine interface passivation with a selective tunneling contact, thus avoiding any need to locally open contact holes in the passivation layer in a real device.