Multiband optical absorption controlled by lattice strain in thin-film LaCrO3

The absorption of light by materials is one of the major steps in converting light energy into electrical energy. The Sun is abundant in visible light and being able to convert sunlight into electricity leads to a free, clean energy source that leaves no carbon footprint. Such energy sources are essential to a safe, secure, and environmentally friendly energy future.

Materials that are of current interest in photovoltaics often include elements that are toxic or rare, such as gallium, arsenic, cadmium, and tellurium. Moreover, the surfaces of these materials react with oxygen in the atmosphere and form oxides, which changes their properties in ways that make them less useful for photovoltaic technologies. A more desirable class of materials for future photovoltaic applications is the metal oxides. These materials can be made from abundant, inexpensive elements, and are stable in air because they are already oxides. In addition, the properties of metal oxides depend strongly on the types of the metal atoms, which can help to engineer oxides with desirable properties.

However, the optical properties of most complex oxides are themselves rather complex (see Figure above), and poorly understood. Gaining a detailed understanding of one such oxide, LaCrO₃, is the focus of this study.

To this end, we have used molecular beam epitaxy method to grow strained epitaxial LCO films on several substrates and analysed their optical absorption properties using spectroscopic techniques (see Figure to the right). In parallel, we carried out a detailed set of theoretical calculations in which we simulated the LCO optical absorption spectra for each strain state (see Figure below). Doing so allowed us to determine in detail which electronic states were involved in absorbing light at specific energies.

What we learned was quite surprising. Earlier experimental investigations led to the conclusion that the onset of electrical conductivity occurs for a light energy of ~3.3 eV, which were thought to excited O-Cr transitions. Our combined experimental and theoretical investigation showed that such transitions occur at a much higher light energy, ~4.8 eV. The absorption features at lower energies with the band maxima at 2.7 and 3.6 eV are due to localised intra-Cr excitations that do not result in electricity being conducted across the LCO, while a band peaking at ~4.4 eV is due to inter-Cr transitions which were disregarded in earlier studies.

In addition, we learnt that the onset of the O-Cr optical transitions can be shifted by ~0.2 eV depending on the substrate, which offers a possibility of tuning optical properties in heterostructures of complex oxides.