water photoelectrolysis assisted by an internal potential – PHOTO-POT

Solar light is intermittent and supply is not synchronized with demand. One way to store solar energy is to use photo-electrolysis of water where hydrogen is produced directly from photons and water in one step. This reaction can be decomposed in three major steps: photon absorption, migration of the photogenerated carriers and surface reaction of these carriers with water. Several metal oxide semiconductors are able to split water into hydrogen and oxygen but with very low efficiencies. The common route to improve metal oxide as photoanode consists in nanostructuring to increase the contact surface with electrolyte and / or in doping to adjust its band gap. In addition a major limitation of oxide layers comes from the high recombination rate of photogenerated electron –hole pairs. In the PHOTO-POT project, which is a fundamental research project both experimental and theoretical,we propose to use the spontaneous electric field present in ferroelectric compounds to improve the performance of photoanodes. Following the “perovskite fever” in the community of solar cell, several groups suggested that using the internal potential of ferroelectric in the field of solar water splitting would greatly improve the efficiency. However, up to now there are very few experimental results. For the present project, we will study model samples: thin, epitaxial films prepared by atomic oxygen plasma assisted molecular beam epitaxy. We have already shown that this method is efficient in the case of hematite to better understand the relevant parameters to improve photoelectrolysis. Specifically, we will grow thin epitaxial films of BaTiO3, a ferroelectric material. Since BaTiO3 has not an optimal band gap for photoelectrolysis, we will dope the film to increase the absorption in the visible range. We will also introduce the BaTiO3 film in a heterojunction where the second layer has a suitable band gap for solar absorption (like Fe2O3). With this panel of samples, the goal is to obtain samples spontaneously upward or downward polarized as well as samples in the polydomain phase. Moreover, we will switch the as-grown polarization by applying external bias prior to experimental characterization when necessary. Therefore these samples will allow us to determine independently the influence of different parameters (orientation or intensity of electric polarization, influence of charges screening) on photoelectrochemical properties. Our preliminary results are promising since we have observed a high increase of the photocurrent (5-fold) when the electric polarization of the ferroelectric is directed towards the substrate. However, many questions remain about the role of polarization:
-Does it improve the photogenerated charge separation?
-Does it increase the reaction rate of the carriers with the water?
-What is the role of ferroelectric domains and domain walls?
-What is the role of screening charges?
We plan to study different model samples in different ferroelectric configurations in order to understand in detail the mechanisms. We will use photoelectrochemical techniques (measurement of photocurrent, impedance spectroscopy and intensity modulated photocurrent spectroscopy) and state-of-the-art photoemission spectroscopy techniques using synchrotron radiation (measurement of the lifetime of the electron-hole pair and of the electronic structure).The experimental results will be compared with DFT calculations to determine electronic characteristics and efficiency for photoelectrolysis. Within this framework, the association of the three laboratories (SPEC, UB-ICB and SOLEIL synchrotron) appears as a complementary and suitable consortium, moreover the partners have already worked and published together for many years. The challenge of this project is to show that internal potential in photoanodes opens a new route for photo-electrochemical cells.