“This is a great example of 3D heteroepitaxy,” Ramamoorthy Ramesh of the University of Maryland told nanotechweb.org. “That is, the two phases are epitaxial with respect to the substrate, but they are also epitaxial in the vertical direction with respect to each other - this leads to epitaxial nanopillars of one phase embedded in the other. The mechanical coupling is much stronger due to this epitaxial matching.”

Ramesh and colleagues made the nanocomposite by pulsed laser deposition from a 0.65BaTiO3-0.35CoFe2O4 target onto an SrTiO3 substrate. Arrays of CoFe2O4 nanopillars with both diameters and average spacing of about 20-30 nm formed in a BaTiO3 matrix. Earlier work on CoFe2O4-BaTiO3 used eutectically solidified materials.

The team carried out magnetic measurements on the material - a change in magnetization at the ferroelectric Curie temperature demonstrated the coupling between the ferroelectric and magnetic order parameters. According to the scientists, thermodynamic analysis showed that the magnetoelectric coupling in the nanostructure is due to strong elastic interactions between the two phases. Coupling between ferroelectric and magnetic order parameters facilitates the interconversion of energies stored in electric and magnetic fields. The material could have applications in devices such as energy transducers (perhaps in data storage) and field sensors.

“The most important thing is that this is not limited to just this combination of perovskites and spinels; there are a lot of systems that do or will exhibit such a self-assembly,” added Ramesh. “So this could be a pervasive approach.”

The team is now trying to understand why and how the nanostructures form, attempting to grow the structures onto silicon, examining other systems that show such self-assembly and working with theorists to find out how to induce a better degree of order among the nanopillars.

“If, indeed, we can get long-range order among the nanopillars (meaning that they arrange themselves into a lattice with a periodicity of say 50-100 nm),” said Ramesh, “this then becomes an even more exciting approach since we can have magnetically and electrically tunable photonic structures. “We are not too far from this, but we still need to work on it.”

The researchers reported their work in Science.