The refractive index of a substance describes how light bends as it enters the material. Most substances have a positive refractive index, which means that light entering a block of glass at an angle to the surface bends towards the normal. But in 1968 Russian physicist Victor Veselago showed that if both the permeability and permittivity of the material were simultaneously negative, refraction would be negative too. In other words, light entering the material at an angle would be bent on the other side of the normal.

Devices that have exploited this odd effect to date, such as high-resolution "superlenses", have all used negative-refraction materials created artificially in the lab, such as arrangements of copper rings or rods. This is because materials with both negative permeability and permittivity have not been found in nature. But now Andrei Pimenov of the Universität Würzburg along with colleagues at other German institutions has shown that negative refraction can crop up in metal ferromagnets – in other words, in natural materials.

Pimenov's team first suspected that ferromagnets might have a negative refractive index after testing materials consisting of layers of ferromagnets and superconductors last year. They found to their surprise that the materials exhibited weak negative refraction if the superconducting layers were in the normally-conducting phase and if an applied magnetic field was used to keep the ferromagnetic layers in a "resonant" state, which is when the magnetic moments rotate at the same frequency as the incident light. This prompted them to see if pure ferromagnets could behave as negative-refraction materials in their own right.

The team shone light on thin films of the metallic ferromagnet La2/3Ca1/3MnO3 and then measured how the amplitude and phase of the transmitted light changed using an interferometer. Using these values they could calculate the permittivity and the permeability, and hence the refractive index. For frequencies of light up to 150 GHz they found that the refractive index was negative. At higher frequencies, however, the effect began to peter out.

Pimenov told Physics Web that metallic ferromagnets should, in principle, be able to have a negative refractive index up to frequencies of 1 THz. However, he added that such materials are unlikely to have negative refractive index at optical frequencies (above 450 THz), which would rule them out as optical superlenses. Nevertheless, Pimenov said that his team is now going to start testing other materials including iron. These could demonstrate negative refraction while in a resonant state at frequencies slightly above 150 GHz, which could be useful for telecommunications.