Project funded by the Swiss National Science Foundation, project 137868: Reaction-diffusion processes for the growth of patterned structures and architectures: A bottom-up approach for photoelectochemical electrodes.

Metal oxides such as hematite (αFe2O3) and tungsten oxide (mWO3) are suitable photoanode materials for photoelectrochemical water splitting. These oxides are indeed stable in corrosive environments and can absorb light in the visible and near-visible ranges, where the solar power spectrum is maximal. They are typically coated as films on transparent conductive substrates to build a photoanode. Nevertheless they suffer from both a poor absorbance and a poor conductivity. On one hand, their low absorbance means these materials need to be thick in order to convert photons into electrons and holes. On the other hand, their low conductivity means they need to be thin in order to separate the photogenerated electrons and holes and drive them to the solid/liquid interface. This contradiction prohibits the use of flat films and the photoelectrode microstructure and nanostructure have to be controlled to decouple light path inside the photoactive film from its thickness.

The aim of this project is to study original photoelectrode films, with a controlled structure grown using bottom-up approaches. Inexpensive self-organization processes are implemented to build architectures with characteristic sizes in the nanometric to micrometric range. At these length scales photonic light management can increase the light path length inside the photoactive materials.

mWO3/hematite microspheroids for light trapping and water splitting

Implementing a heterojunction between a tungsten oxide thin film (thickness around 300 nm) and a hematite ultra-thin film (thickness around 5 nm) in a photoanode proved to be a successful strategy to observe significant water splitting under simulated solar light. The depletion layer at the interface between mWO3 and hematite improves the photogenerated electron/hole separation. This separation is also improved by decreasing the oxides film thicknesses down to their charge diffusion length. The following animated band diagram shows how charges are generated and separated in a photoanode composed of a mWO3/hematite heterojunction, under operation:

Nevertheless for a good charge separation this design relies on thin oxide films. The films thinness, coupled to the transparency of metal oxides, leads to poor light absorption. In our article: Photonic light trapping in self-organized all-oxide microspheroids impacts photoelectrochemical water splitting (Energy and Environmental Science, Authors version) we present a bottom-up approach to implement this kind of heterojunction into microspheroids and we explain how light is trapped by this architecture and increases the film photoactivity. This structure is self-organized, emerging from a vesicle-templated sol-gel process. Polymer vesicles self-assemble in an aqueous solution containing a soluble tungsten precursor. Upon spin coating of this solution a self-organized monolayer of tungsten sol droplets embed in a polymer matrix is formed. When this film is pyrolysed mWO3, the tungsten sol is converted to mWO3 while the polymer is decomposed, forming the microspheroids. An iron sol is then coated on the spheroids and pyrolysed to form an ultra-thin hematite overlay (5 nm). The following figure shows a SEM image of the resulting microspheroids on the transparent conducting substrate (side view):

Due to their shape and mesoscale dimensions light is trapped in the film. A simulation of light interaction with the microspheroids was performed and confirmed experimentally. We observed two different photonic effects: resonant confined modes inside the microspheroids and near-field scattering. You can see these two effects in the following video that shows a simulated comparison of light interaction with a flat film and a microspheroid:

You can observe in this video that light is deviated from normal incidence and trapped inside the microspheroid (resonant modes) while part of this trapped light is re-emitted outside the spheroid and can be caught again by neighboring spheroids (near-field scattering). In our article (Energy and Environmental Science, Authors version) we showed that these light trapping effects increases the film photoactivity compared to flat films and we investigated how the spheroids sizes impacts the film photonics and photoactivity.

Electrohydrodynamic lithography to structure hematite

We also investigated an alternative self-organization method: electrohydrodynamic lithography (EHD). A polymer/iron salt thin film is produced by spin coating of an aqueous solution on a silicon wafer. This film is submitted to a potential difference while facing a nanostructured counter electrode. Using this method, hematite films with a hierarchical structure were obtained:

This complex structure is composed of micrometric oblate droplets (0.5-1 μm) covered with nanometric oblate droplets (100-200 nm), the overall film being composed of nanoparticles (20-30 nm). This work constituted our first successful implementation of a complex electrode architecture involving a functional photoelectrode material, hematite.

For more information you can consult our article: Hematite nanostructuring using electrohydrodynamic destabilization (Applied Surface Science, Authors version).

Publications

Public communications

Collaborators

	Partner Institutions	Research Team
	Empa Swiss Federal Laboratory for Materials Science and Technology	Florent Boudoire, Phd Student Artur Braun, PI Jakob Heier, PI Rita Toth, Scientist
	University of Basel	Florent Boudoire, Phd Student Edwin C. Constable, PI

• Project description
• WO3/hematite microspheroids
• Hematite nanostructuring
• Publications
• Public communications
• Collaborators

Funding agency:

Collaborating institutions: