Photoelectrochemical Hydrogen Production - Final Report

The scope of this photoelectrochemical hydrogen research project is defined by multijunction photoelectrode concepts for solar-powered water splitting, with the goal of efficient, stable, and economic operation. From an initial selection of several planar photoelectrode designs, the Hybrid Photoelectrode (HPE) has been identified as the most promising candidate technology. This photoelectrode consists of a photoelectrochemical (PEC) junction and a solid-state photovoltaic (PV) junction. Immersed in aqueous electrolyte and exposed to sunlight, these two junctions provide the necessary voltage to split water into hydrogen and oxygen gas. The efficiency of the conversion process is determined by the performance of the PEC- and the PV-junctions and on their spectral match. Based on their stability and cost effectiveness, iron oxide (Fe2O3) and tungsten oxide (WO3) films have been studied and developed as candidate semiconductor materials for the PEC junction (photoanode). High-temperature synthesis methods, as reported for some high-performance metal oxides, have been found incompatible with multijunction device fabrication. A low-temperature reactive sputtering process has been developed instead. In the parameter space investigated so far, the optoelectronic properties of WO3 films were superior to those of Fe2O3 films, which showed high recombination of photo-generated carriers. For the PV-junction, amorphous-silicon-based multijunction devices have been studied. Tandem junctions were preferred over triple junctions for better stability and spectral matching with the PEC junction. Based on a tandem a-SiGe/a-SiGe device and a tungsten trioxide film, a prototype hybrid photoelectrode has been demonstrated at 0.7% solar-to-hydrogen (STH) conversion efficiency. The PEC junction performance has been identified as the most critical element for higher-efficiency devices. Research into sputter-deposited tungsten trioxide films has yielded samples with higher photocurrents of up to 1.3 mA/cm2. An improved a-Si/aSi tandem device has been demonstrated that would provide a better voltage match to the recently improved WO3 films. For a hybrid photoelectrode based on these component devices the projected STH efficiency is 1.3%. For significant efficiency enhancements, metal oxide films with increased optical absorption, thus lower bandgap, are necessary. Initial experiments were successful in lowering the WO3 bandgap by nitrogen doping, from 3.0 eV to 2.1 eV. Optimizing the electronic properties of these compounds, or other reduced-bandgap materials such as Fe2O3, is the most immediate challenge. As the photocurrent levels of the PEC junction are improved, increasing attention will have to be paid to the matching PV junction