Tungsten Trioxide Photoanode-Palladium Cathode Cell for Solar Seawater Splitting

Abstract

N-type semiconductor-tungsten trioxide (WO3) finds application in a large variety of fields including photocatalysis, gas sensors, smart windows or as a substrate for heterogeneous catalysts. This presentation will focus on recent improvements in optoelectronic and photoelectrochemical (PEC) properties of WO3 thin films achieved in our laboratory. The semitransparent WO3 films are formed on F-doped tin oxide (FTO) conductive glass substrates, using a sol-gel method, by depositing the precursor layer-by-layer and annealing sequentially in the flow of oxygen. Use of appropriate mixtures of tungstic acid and organic structure-directing agents that form the WO3 film precursor, combined with a two-step annealing, allows the formation of highly crystalline nanostructured (NS) electrodes with controlled, large porosity. This renders possible the permeation of the whole mesoporous film by the electrolyte. Annealing of the as-deposited samples in oxygen above 500°C ensures structural ordering with formation of monoclinic WO3. Since none among photostable (metallic oxide) semiconductors exhibits band-edge energy levels that match those of hydrogen and oxygen evolution reactions to allow unassisted water splitting, the present efforts focus on minimizing the bias voltage required to perform visible light-driven photooxidation of water. In fact, a bias voltage of the order of 1 V may be provided by several single-junction PV cells likely to operate in a tandem device with the PEC cell. Despite visible-light absorption range of the WO3 films restricted to 500 nm, under standard conditions, i.e. at 1.23 V vs RHE and under simulated AM 1.5G solar light (100 mW/cm2), stable anodic water splitting photocurrents exceeding 4.5 mA/cm2 are regularly attained.This presentation will principally focus on photoelectrolysis experiments employing synthetic seawater electrolyte, in a two compartment (separated by a sintered glass diaphragm) cell, that showed dominant (with 60-70% Faradaic efficiency) formation of chlorine at the WO3 photoanode with remarkably stable photocurrents reaching 4.7 mA/cm2. In fact, as demonstrated in an earlier work from our laboratory1, the WO3 photoelectrodes exhibit preferentially oxidation of anions (including Cl-) of the acidic aqueous electrolytes. However, in contrast with the Cl- ions, the photooxidation of oxy-anions of acidic electrolytes leads generally to the "passivation" of the WO3 electrodes due to the formation of the surface layers of peroxo species.1 The only identified exception is the methane-sulfonic acid that allows oxygen generation at the WO3 photoanode with large stable photocurrents.2 In the case of seawater photoelectrolysis, the operating potential point at which the photocurrent approaches saturation occurs at ca 0.4 V below E(Cl-/Cl2)=1.36 V vs SHE. Consistently, the incident photon conversion efficiencies (IPCEs) reach a maximum of 85% over 390-410 nm range of wavelengths.3 Good mechanical and chemical stability of the WO3 electrodes in the presence of chlorine evolved from seawater was confirmed by monitoring the photocurrents over a 20 hours-long continuous photoelectrolysis runs.A schematic diagram of the seawater splitting PEC cell, including a NS WO3 film photoanode and a Pd sheet cathode, used in our experiments is shown below.Fig. 1 Schematic diagram of the seawater splitting PEC cell.The current of CO2 gas passing through the seawater represents a possible option for maintaining slightly acidic pH in the cathodic cell compartment. The important advantage of the adopted PEC cell configuration is to circumvent the problem of gas separation, i.e. chlorine/oxygen formed at the photoanode and hydrogen generated at the Pd cathode that is stored in the form of a Pd-H hydride. The storage of hydrogen generated in a PEC water splitting device in the form of a hydride, opens the possibility of its subsequent use as an anode in a hydride-oxygen battery.Such sunlight-driven PEC seawater splitting device can be implemented in remote places to provide on-site relatively small amounts of chlorine for disinfecting purposes including purification of drinking water.More generally, the prospect of using seawater electrolyte for a larger scale PEC hydrogen generation would release the strain on vital fresh water resources. Considering relatively low concentration of Cl- ions in seawater, roughly equivalent to 0.5 M NaCl, the photoelectrolysis delivering photocurrents in the range of mA/cm2 appears well suited for such applications.