The performance of thin films in display devices, smart windows, self-cleaning coatings, and solar cells depends strongly on the crystal structure, microstructure, and chemical composition of the host materials. The sources of heterogeneity resulting from existing technologies can be identified by scrutinizing the film-formation processes. Populations of different intermediate chemical clusters produced in gas phase or solution are deposited on a substrate. These populations exhibit different reactivities, making control over nucleation, growth, coalescence, and mass-transfer processes very difficult. This is a big problem that materials scientists have tried to solve. Analytical techniques show films consisting of crystallites (‘grains’) of different sizes, shapes, and structures. Structural and chemical defects inside the crystallites and at their boundaries are quenched during aggregation and packing. This heterogeneity affects, among others, charge-carrier mobility and ion diffusion.
Figure 1. (top left) High-resolution transmission-electron-microscopy (HRTEM) image of a new tungsten trioxide hexagonal structure. (top right) The same image after Fourier filtering and a structural model for the new metastable hexagonal phase. (bottom left) A model for oxide clusters made of three tungsten atoms (tritungstic groups) and their condensation process.
To overcome this problem, a promising approach was developed by generating a single cluster, controlling its atomic packing to match that of the host material's crystal structure, and then using it as a single source to grow films. The tendency of these clusters to self-assemble and better pack leads to ordered and relatively stable structures. This enables us to synthesize homogeneous films of different packing orders. From a fundamental point of view, this ‘single-cluster-source approach’ also allows a systematic study of the relationship between a material's structure and its properties.
We recently reported the first use of tritungstic groups as single-cluster source to process homogeneous films. We successfully used high-resolution transmission electron microscopy (HRTEM) to image the atomic arrangement in the resulting homogeneous electrochromic film. A new hexagonal phase was isolated. This open-host structure explains the good lithium-intercalation reaction and the excellent electrochromic response of tungsten trioxide films. Figure 1 shows an HRTEM image of the structure, while the scheme at the bottom illustrates the mechanism by which tritungstic groups self-assemble and condense, thus leading to the pure phase.1
Inspired by theoretical work suggesting dimers are intermediate building units in the formation of titania-anatase crystal structure,2 we developed a chemical route to stabilize and isolate pure oxide clusters made of two titanium atoms (dimers). Highly homogeneous, transparent, and crack-free films are made using these chemical compounds as single source. The photoluminescence spectra at different temperatures show only excitonic luminescence and no defect luminescence. Investigation of dye degradation shows that these films are highly photocatalytically active. They can be used as self-cleaning coatings or in photoreactors for water purification. Dimers are also used to generate larger clusters or nanocrystals. Some of these nanocrystals tend to self-assemble, forming a variety of highly nanostructured films (see Figure 2).
Figure 2. HRTEM images of two types of nanostructured titania films resulting from the single-cluster-source approach.
A high-pressure titania phase was attained at ambient pressure by differently capped clusters used for the processing of titania films. One very important aspect of this chemistry is that small clusters are made as preformed building blocks for phases generally made through high-temperature solid-state reactions. Water-soluble titanate clusters can be made with the same atomic packing as the building blocks in oxide clusters of six titanium atoms (hexamers), a phase usually made at temperatures higher than 1000°C. This solubility is a great advantage for building multicomponent systems with applications in catalysis and photocatalysis. Efforts to build multicomponent photocatalysts for water splitting and hydrogen production are under way.3
Helmholtz-Zentrum für Materialien und Energie GmbH
Abdelkrim Chemseddine was educated at the University Pierre et Marie Curie, Paris VI, France. He obtained an MSc in spectrochemistry and structure in 1981, a ‘doctorat de 3mecycle’ in the chemistry of isopolytungstates in 1983, and a doctorat d'état in the sol-gel chemistry of tungsten oxide in 1986. He was subsequently appointed as research associate at the Materials Science and Engineering Department of the University of California, Los Angeles (1986–1987). As an Alexander von Humboldt Foundation guest scientist he worked at the Ludwig Maximilian University in Munich, Germany, between 1987 and 1989. He spent one year teaching chemistry at the Faculty of Science of the university Hassa II, Morocco. As a guest scientist he worked at the Hahn-Meitner Institute in Berlin, Germany, on sol-gel processing of highly porous semiconductor-oxide films. Since 1991 he has been a permanent staff scientist with research interests in the field of nanocrystals. In 1997 he spent four months as a visiting professor teaching the chemistry of nanocrystals and nanostructured materials at the Georgia Institute of Technology.