It sounds too good to be true: high-quality flat lenses that focus light and can be made in sheets and cut to size. That's the promise of a new class of materials with a negative refractive index that bend light in the opposite direction from conventional materials [see diagram].

There's a problem, however. These materials are hard to make. The first ones were made in 1999, and the only ones created to date operate in the microwave region and consist of a complex assembly of copper rings and wires that must be painstakingly assembled by hand into a periodic structure [see "Left-Handed Material Reacts to 3-D Light," IEEE Spectrum, October 2002]. Such materials are not mass-producible using known techniques.

That could be about to change. According to Tom Mackay, a mathematician at the University of Edinburgh, and Akhlesh Lakhtakia, a mechanical engineer at Pennsylvania State University, in University Park, another way to make them is to take a handful of one material and mix it with another. Their breakthrough is the discovery that blended materials can have properties that were not present in the originals.

The possibility of negative refraction was first raised in the late 1960s by the Russian physicist Victor Veselago of the General Physics Institute of the Russian Academy of Sciences in Moscow. Veselago based his ideas on the discovery by James Clerk Maxwell that light's behavior in a material could be described in terms of two constants: for the magnetic field, the material's permeability, and for the electric field, its permittivity.

The permittivity and permeability of all natural materials have positive values. Veselago asked what would happen if he switched the sign of these constants from positive to negative, and found that the material would bend light in the opposite direction it normally is refracted.

Taking Veselago's thought experiments a step further, Mackay and Lakhtakia asked whether ordinary materials could ever combine in a way that produces negative permittivity and permeability. To their surprise, the equations showed that this is possible using easily available materials. "One material needs to be metallic in character while the other must be magnetic," says Mackay. "It was really a very unexpected result."

There are a couple of other conditions. The materials must be formed as spheres that are about the same size as the wavelength of light they are designed to bend. And the materials must be mixed in certain proportions. Mackay says that the technique should make it possible to bend visible light the wrong way for the first time.

"Nanoparticles of the size needed to manipulate visible light are now common," he says. "There is nothing to prevent this from working in the optical regime." Until now, however, the technique has worked only with microwave.