LBNL Masthead A-Z IndexBerkeley Lab mastheadU.S. Department of Energy logoPhone BookJobsSearch
Beyond the Lone-Pair Model for Structurally Distorted Metal Oxides Print
Wednesday, 28 February 2007 00:00


"Ferroelectricity," by analogy to ferromagnetism, is defined as the presence of spontaneous electrical polarization in a material, often arising from distortions in the material's crystal structure. In oxides of the metals lead and bismuth, such distortions were for many years attributed to the existence of "lone pair" electrons: pairs of chemically inert, nonbonding valence electrons in hybrid orbitals that leave noticeable voids in the crystal structure. At the ALS, researchers from the U.K., Ireland, and the U.S. have now obtained definitive experimental evidence that this lone-pair model must be revised. High-resolution x-ray photoemission spectroscopy (XPS) and soft x-ray emission spectroscopy (XES) have clarified the subtle electronic origins of the prototypical distortions in these crystal structures. The results have important implications for the tantalizing possibility of spintronic or superconducting devices combining ferroelectric and ferromagnetic properties.

Flawed but Interesting

Perfect symmetry and regularity can be beautiful, but boring. In some materials, it is the slight imperfections found on the atomic scale that can lead to the most interesting—and potentially valuable—macroscopic properties. For example, chromium impurities are what give rubies and emeralds their brilliant color. In semiconductors, the introduction of impurity atoms, or dopants, provides the extra electrons or holes needed to conduct electricity. Ferromagnetism occurs when unpaired "extra" electrons in incompletely filled shells are free to align their spins in a particular direction. And, as discussed here, ferroelectricity arises in materials whose crystal structures have been nudged a bit out of alignment, creating an asymmetric charge distribution (i.e., polarization) inherent in the bulk material.

The coupling of the latter two phenomena, ferromagnetism and ferroelectricity, in composite materials or even in a single "multiferroic" material, offers the exciting prospect of devices in which electric polarization can be induced by a magnetic field and magnetization can be induced by an electric field—providing an extra "degree of freedom" for device designers. As we get better at synthesizing complex nanostructures that mix and match these properties, studies such as this one by Payne et al. become increasingly important in helping us to understand and exploit the possibilities inherent in Nature's imperfections.

Top: In the low-temperature phase of lead oxide (α-PbO), each lead atom (blue) forms a tetragon with four oxygen nearest neighbors (red) on one side. Bottom: In monoclinic bismuth oxide (α-Bi2O3), the two distinct bismuth ions (blue) both have five oxygen near neighbors (red), creating a distorted square pyramidal coordination geometry.

The asymmetrical crystal structures of the metal oxides α-PbO and α-Bi2O3 were for many years explained by the hybridization of the metals' 6s and 6p orbitals, which are occupied by metal 6s electrons, assumed to lie close to the Fermi energy (EF) in the solid state. This conventional "lone-pair" model, however, has recently been called into question on the basis of density functional theory (DFT) calculations that suggest that the majority of the 6s population in α-PbO is in fact found at the bottom of the main valence band, about 10 eV below EF. However, definitive experimental evidence supporting this idea has been lacking.

In this study, the investigators showed that consideration of the relative intensities of valence-band components in oxygen K-shell XES, obtained at ALS Beamline 7.0.1, provides a simple but incisive experimental approach to investigating the nature of lone-pair states in metal oxides, especially when compared with the intensities of features in XPS obtained from an aluminum Kα x-ray source. Both XPS and XES spectra (and the corresponding DFT calculations) show a well-defined band (labelled III) that lies about 10 eV below the top of the valence band. However, this band diminishes dramatically in the XES data. Since oxygen K-shell XES is governed by a very strict selection rule that allows only states of oxygen 2p character to decay into the oxygen 1s core hole, the XES data directly measures the oxygen 2p partial density of states (PDOS).

XPS (top left) and XES (bottom left) spectra of α-PbO compared to DFT calculations of the total density of states (DOS, top right) and partial density of states (PDOS, bottom right) with the contributions from the various O and Pb states indicated. The XPS data correspond to the total DOS while XES measures the O 2p PDOS. The spectra are all presented on a binding energy scale referenced to the top of the valence band. (The results for α-Bi2O3 are similar.)

The conclusion, fully supported by DFT calculations, is that there can be little oxygen 2p character in this band, and by default, the associated electronic states must have dominant metal 6s character. These findings confirm that the structural distortions in α-PbO and α-Bi2O3 should not be attributed to the direct hybridization of metal orbitals close to EF, resulting in purely metal-based 6s-6p lone pairs. Instead, the dominant contribution to the metal 6s PDOS is found at the bottom rather than the top of the valence band, and indirect mixing between 6s and 6p states is mediated by hybridization with oxygen 2p states at the top of the valence band.

It follows that qualitative textbook explanations of structural distortions in many heavy post-transition-metal compounds should be revised, with important implications for understanding the structural physics of, for example, magnetic ferroelectric materials such as BiMnO3 and BiFeO3. The results are also of general significance in relation to the electronic structures of ternary lead and bismuth oxides, including the many high-temperature superconducting phases that contain these heavy cations. It has in the past been assumed that the 6s states lie close to EF and therefore contribute significantly to the states responsible for conduction in metallic phases. The present findings demonstrate that this viewpoint is not correct. Again, this will have an impact on our understanding and tuning of the physical properties of these and related materials.



Research conducted by D.J. Payne and R.G. Egdell (Oxford University); A. Walsh and G.W. Watson (Trinity College); J. Guo (ALS); and P.-A. Glans, T. Learmonth, and K.E. Smith (Boston University).

Research funding: U.K. Engineering and Physical Sciences Research Council; Higher Education Authority (Ireland); U.S. Department of Energy, Office of Basic Energy Sciences (BES). Operation of the ALS is supported by BES.

Publication about this research: D.J. Payne, R.G. Egdell, A. Walsh, G.W. Watson, J. Guo, P.-A. Glans, T. Learmonth, and K.E. Smith, "Electronic origins of structural distortions in post-transition metal oxides: Experimental and theoretical evidence for a revision of the lone pair model," Phys. Rev. Lett. 96, 157403 (2006).