The relative stabilities of different transition metal oxide phases — such as Co3O4 versus CoO or the range of iron oxides — have been extensively studied and are well understood in bulk materials. As scientists have shrunk the size of those materials down to the nanoscale, however, they have found that these stabilities are often very different. Although this has been attributed to surface energy effects, which have a much greater influence on such small volumes, this is not fully understood.

Now, Alexandra Navrotsky and colleagues from the University of California at Davis have studied1 the stability of a range of metal oxide systems at the nanoscale. They discovered that surface energies profoundly affect the equilibria between different redox states and thus the stability of these phases. Using calorimetric methods, Navrotsky and co-workers determined that the spinel Co3O4 phase is greatly stabilized in 10-nm nanoparticles compared with rock salt CoO or the pure metal. Comparison with oxides of other metals — iron, manganese and aluminium — suggest that more oxidized phases may be more stable in general.

Understanding these relationships can help to explain observed nanoparticle behaviour; for example, the ability of nanosized Co3O4 to oxidize carbon monoxide at high CO/CO2 ratios with low oxygen supply. Furthermore, a calculated phase diagram for nanosized iron oxides shows no stable region for Fe0.947O, tallying with observations of their spontaneous oxidation to form materials useful for memory applications.