Photoelectrochemical hydrogen production from aqueous solution employing nanostructured semiconductors

Abstract

The focus of this thesis is the exploitation of various types of photoelectrodes based on nanostructured materials to be employed in photoassisted electrolysis schemes, aimed to convert solar energy into chemical energy. Since the world demand of energy is projected to increase in the next years, environmental issues like global warming and climate change impose a change in our way of producing and storing energy, currently and mainly based on the combustion of fossil fuels. Both a stable improvement of the quality of life for an increasing world population and the preservation the environment will require the use of progressively larger amounts of new energy sources which must be carbon neutral, renewable and low cost. This defines a grand scientific and technological challenge for the XXI century. Sunlight provides by far the largest of all carbon-neutral energy sources. More energy from sunlight strikes the Earth in one hour (4.3 x 1020 ) than all the energy consumed on the planet in a year (4.1 x 1020 J). Yet solar electricity provides only approximately 1 millionth of the total electricity supply, and renewable biomass provides less than 0.1% of the total energy consumed. There is a huge gap between our present use of solar energy and its enormous undeveloped potential. Hydrogen production from solar water splitting is regarded by many as the “holy grail” of photochemistry, providing a practically inexhaustible fuel source and, since the experiment of Fujishima and Honda, the use of semiconductor photoelectrodes capable of operating with a virtually unitary quantum yield has appeared as a convenient mean to realize water splitting in a photoelectrochemical cell. In particular, a large part of this thesis work has been devoted to the development of efficient WO3 photoelectrodes which is a promising material for water splitting, coupling favourable energetics for oxygen evolution, high photochemical stability in aqueous media and visible photon absorption up to 500 nm. . At first, different synthetic approaches for obtaining nanostructured WO3 on transparent conductive oxides have been explored by keeping the Santato-Augustynski recipe as a main guideline. In general the preparation of nanostructured photoelectrodes involves the formation of stabilized H2WO4 colloidal precursors in the presence of different organic dispersing agents which control both nanoparticle size and film porosity. The relationships between crystallite size, film porosity, electrochemically active surface and photoelectrochemical behavior under simulated solar illumination have been studied. The work has then focused on the preparation of WO3 films by potentiostatic anodization of metallic tungsten in different solvent/electrolyte compositions with the aim of improving the charge transfer and charge transport kinetics by creating tightly interconnected porous oxide structures directly bound to a metal collector. In the NMF/H2O/NH4F solvent mixture the anodization leads to highly efficient WO3 photoanodes, which, combining spectral sensitivity, high electrochemically active surface and improved charge transfer kinetics, outperform, under simulated solar illumination, most of the reported nanocrystalline substrates produced by anodization in aqueous electrolytes and by sol gel methods. The use of such electrodes results in high water electrolysis yield, of the order of 70% in 1 M H2SO4 under a potential bias of 1 V vs SCE and close to 100% in the presence of methanol acting as a hole scavenger. The requirement to harvest large part of the solar spectrum and photooxidize small molecules like H2S (largely present as a contaminant of hydrocarbon sources) has led us to the study of heterointerfaces between TiO2 and group VI semiconductors like CdS, CdSe, CdTe and Bi2S3, which are characterized by a relatively narrow band gap (from 2.4 eV for CdS to 1.4 eV for Bi2S3). The use of colloidal and nanotubular TiO2 as a substrate for growing sensitizing semiconductors by chemical and electrochemical means was instrumental in reducing significantly degradative self photoxidation and recombination leading to good photon to electron quantum yields, ranging from 20% (Bi2S3) to 60% (CdS) The photoelectrochemical behavior of TiO2 photoanodes sensitized by water stable molecular Ru(II) dyes was finally explored in the context of photoinduced hydrogen generation. The photoanodes sustained 240 h of irradiation without undergoing appreciable hydrolysis and decomposition in an aqueous environment at pH 3. Despite a favourable driving force for direct water oxidation, the performances in pure water were far from an applicative interest. However, the use of organic sacrificial donors, like isoprapanol and ascorbic acid considerably enhanced their photoanodic response and, interestingly, the exploitation of iodide, a well known electron donor in photoelectrchemical cells, was more problematic because the adsorption of photogenerated I3 - from aqueous media favored charge recombination with conduction band electrons, thus limiting the efficiency of the photoelectrosynthetic device. However, experiments performed in a three-compartment cell, where the photolectrode was in contact with an organic solvent preventing triiodide adsorption, revealed a remarkable photocurrent, with an electrolysis yield close to 87%. Although the problem of finding the ideal system for photoassisted electrolysis has not been entirely solved, this thesis work suggests some feasible approaches for obtaining high photon to electron conversion efficiencies, pointing out that the basis of an efficient photoelectrochemical system reside in the combination of an high electroactive area with fast interfacial charge transfer kinetics, necessary to overcome losses arising from recombination.