Schottky Tunnel Contacts for Efficient Coupling of Photovoltaics and Catalysts

This interdisciplinary project investigates the performance of nanoscale metalinsulator-semiconductor (MIS) contact structures that electrically couple optimized catalysts to high quality semiconductor absorbers in photoelectrochemical (PEC) cells, while chemically protecting the absorbers from oxidation during solar-driven water splitting. In the 2013-14 funding period, we completed and published a study on water oxidation performance of varying the TiO2 thickness (grown by atomic layer deposition – ALD) in the MIS structures.1 Uniform films of atomic layer deposited TiO2 are prepared in the thickness range ~1-12 nm on degenerately-doped p+-Si yielding water oxidation overpotentials at 1 mA/cm2 of 300 mV to 600 mV in aqueous solution (pH 0 to 14). Electron/hole transport through Schottky tunnel junction structures of varying TiO2 thickness was studied using the reversible redox couple ferri/ferrocyanide. The dependence of the water oxidation overpotential on ALD-TiO2 thickness, with all other anode design features unchanged, exhibits a linear trend corresponding to ~21 mV of added overpotential at 1 mA/cm2 per nanometer of TiO2 for TiO2 thicknesses greater than ~ 2 nm. This linear behavior for anodes with thicker TiO2 layers suggests bulk-limited electronic conduction through the insulator stack. A model of the field needed for bulklimited transport at the given current densities agreed quantitatively with the experimental results lending further evidence for this hypothesis. Additionally, we have shown that ALD-TiO2 protection can be used with three times less iridium with only small effects on device performance as well as with a variety of other metal catalysts.1 These findings together constitute a major advance in the coupling of general photovoltaic and catalyst systems. In light of our findings of the bulk-limited conduction through thicker TiO2 oxide layers, we recently performed another set of experiments varying the SiO2 oxide thickness. An understanding of the tunnel oxide SiO2 layer in coordination with an understanding of the ALD-TiO2 layer will allow for better control of the overall stack conductivity and stability. We have also assembled a new ALD reactor to investigate precursor chemistries for deposition of RuO2-TiO2 alloy layers that combine the capabilities of an efficient oxygen evolution reaction (OER) catalyst and a chemical protection layer for an underlying silicon absorber. Our goal is to minimize utilization of RuO2, a rare, but very efficient,