Crystalline-Silicon/Organic Heterojunctions for Solar Photovoltaics

Abstract

Solar cells based on crystalline silicon offer high efficiency but they are expensive due to the high temperatures required in their fabrication. The alternative approach using low-temperature processable organic-semiconductors is potentially cheaper, but the organic solar cells are not very efficient. In this thesis we explore if organic semiconductors can be integrated with silicon to form hybrid organic/silicon solar cells that are both efficient and low-cost. Specifically, we demonstrate that a) organic molecules can be used to reduce carrier recombination at the silicon (100) surface and b) a solution-processed organic/silicon heterojunction can replace the conventional silicon p-n junction to yield solar cells with high power conversion efficiencies (> 10 %).

With decreasing wafer thicknesses and improving bulk lifetimes of silicon solar cells, losses due to carrier recombination at the silicon surface are becoming increasingly important. At a bare silicon surface, some of the silicon valencies remain unsatisfied. These ``dangling-bonds'' cause midgap states at the silicon surface where photogenerated carriers can recombine, resulting in lower performance. Typically, a layer of silicon oxide/nitride is deposited on the silicon, at high-temperatures (> 350 &deg; C), to passivate the dangling-bonds and reduce surface recombination. Organic semiconductors can be deposited at much lower temperatures, but in general organic materials do not react with the silicon dangling-bonds and the surface remains unpassivated. In this work, we demonstrate that the organic molecule 9,10 phenanthrenequinone (PQ) reacts with and satisfies the silicon dangling bonds, leading to a relatively defect-free silicon surface with a very low surface recombination velocity (~150 cm/s). Electrical measurements of the metal/insulator/silicon devices show that the Fermi-level at the PQ-passivated silicon surface can be modulated and an inversion layer can be induced in silicon. High electron mobility of 600 cm<super>2</super>/Vs is measured at the Si/PQ interface further proving the electronic quality of the PQ-passivated surfaces.

To generate a photovoltage in a solar cell, the photogenerated carriers need to be spatially separated at two electrodes of opposite polarity. In solar cells this is typically accomplished using a p-n junction. While the p-n junction technology is well understood, the fabrication of p-n junctions on silicon is an expensive process because it requires ultra-clean furnaces, pure precursors and high temperatures. In this thesis we successfully replace the silicon p-n junction with an silicon/poly(3-hexylthiophene) heterojunction that can be manufactured at low temperatures (< 150 &degC) with a simple spin-coating process. The key design rules to achieve a high quantum-efficiency and high open-circuit voltage are discussed and experimentally demonstrated. Finally we highlight the importance of reducing minority-carrier currents in these heterojunction devices, which gives a pathway for further improving the efficiency of heterojunction solar cells. Using the prescribed design rules and optimizing device structure, a silicon/organic heterojunction solar cell with an open-circuit voltage of 0.59 V and power conversion efficiency of 10.1 % is demonstrated.