Stable and efficient photoelectrodes for solar fuels production

Abstract

[eng] The excessive consumption of non-renewable energy sources such as fossil fuels has lead the world to a global climate change, urging for new energy consumption habits together with developing cost- effective alternative renewable technologies. Photoelectrochemical (PEC) water splitting allows for direct conversion of solar light and water into hydrogen and oxygen, storing energy into chemical bonds, solving the storage problem of photovoltaic technology. It has demonstrated to produce pure hydrogen and oxygen in significant efficiencies, although this technology is not ready for market implementation due to lack of efficient, stable and scalable photoelectrodes. In this work, we undertake a journey from improving the efficiency of stable metal-oxide-based photoanodes to stabilizing efficient photovoltaic materials by the introduction of protective, transparent, conductive and catalytic layers. Efforts have focused on using cost-effective and scalable materials and techniques. Metal oxide candidate TiO2 is reported stable in alkaline electrolytes and at anodic potentials, but they present low photon to current conversion efficiencies. This is due to excessively large band gap, absorbing small part of the visible spectra, and small electron and hole mobility. Its efficiency is increased both by microstructuring the substrate and nanostructuring the thin film into nanorods, and by modifying the electronic structure with a reductive H2 treatment, enhancing potential drop inside the nanorods. The strategy is shifted into stabilizing highly efficient short band gap semiconductor materials used by the photovoltaic industry. Silicon based photocathodes are protected from acidic electrolyte corrosion by TiO2 overlayers grown by atomic layer deposition (ALD). Temperature is found to play a key role for both efficient film conductivity and stability, being this caused by polycrystalline films formation. ALD enabled high thickness control and pinhole-free layers, together with lower crystallization temperatures than other techniques. Copper-indium-gallium-selenide (CIGS) solar cells fabricated on flexible stainless steel substrates are also protected from corrosion by TiO2 ALD protective layers. The transparent conductive oxide (TCO) already used in solar cells is found necessary for efficient p-n junction formation and charge transport to the hydrogen evolution reaction. Copper-zinc-tin- sulfide/selenide (CZTS/Se) solar cells, where scarce indium and gallium are substituted by tin and zinc, are implemented for PEC devices with TiO2 overlayers too. By modifying the S/Se ratio, band gap can be tuned, an especially interesting characteristic to design tandem PEC devices. ALD deposited protective layers are also studied in anodic polarizations and alkaline electrolytes. By varying the deposition temperature of TiO2, completely amorphous, mixed amorphous and crystalline and fully crystalline films are deposited, and a clear conductivity increase is observed correlated to crystallization. Preferential conductivity paths are observed inside crystalline grains, proposed to be related to crystalline defects and grain boundaries. Few hundred hours stability tests reveals significant photocurrent decrease, with no observed dissolution of the Si photoabsorber. This is attributed to oxidative potentials and electrolyte hydroxides diminishing the n-type semiconductor behavior of TiO2 and forming a barrier to charge injection into the oxygen evolution reaction. UV superimposed illumination partially recovered conductivity. NiO films are ALD-deposited on Si photoanodes and

conductivity is found to decrease when temperature is increased from 100 to 300 ºC, simultaneous to a change in preferential crystal growth direction. Higher stoichiometric film, being formed when increasing temperature, decreases Ni2+ vacancies, responsible of the p-type semiconductor behavior. Impressive 1000 hours stability measurements are obtained. Although, this is only attained under periodic cyclic voltammetries, avoiding partial deactivation of the photoanodes. This is attributed to chemical modifications at the surface in such highly oxidative conditions.