PASADENA, Calif. -- You would think that increasing the efficiency of a third-order nonlinear material by 35 times at the commercially attractive wavelength of 1.5 µm would be enough to make a researcher happy. After all, it could be an important step toward all-optical switching in local cross-connect telecommunications networks. Instead, California Institute of Technology chemist Seth Marder and George Stegeman, a physicist at the University of Central Florida's CREOL facility in Orlando, say their latest findings are a means to an end -- not the end itself. "The point was to find a compass with an arrow showing direction," Stegeman explained. Nonlinear materials have several uses; among those purposes is making a hip place for groups of multiple photons to combine and make light of a different wavelength or color. Another interesting characteristic is the materials' ability to change their optical properties, such as the refractive index. This particular attribute excites more than just molecules. Researchers interested in optical networks and computing hope to develop all-optical switches that can change the path of a light beam without an external electrical field. Second-order nonlinear materials generally depend on electrical fields to excite the material at the molecular level and thereby change a laser's properties. In third-order materials, an external light replaces the electrical source -- hence the all-optical switch. Marder and Stegeman's research should help pave the way for new families of third-order materials. Taking a cue from earlier research showing that polarized polyenes (molecules) make good nonlinear optical materials, the researchers altered a molecule to increase its polarization. They started with b-carotene, a chromophore that exhibits the strongest recorded nonlinear response. The molecule itself is a chain of carbon molecules. At one end is a ringlike structure that acts as an electron trap, pulling electrons from the other end of the molecule. As electrons progress to one end of the molecule, it creates a polarized charge. The wider the separation and the higher the difference in the charge, the greater the nonlinear efficiency, Marder said. By switching the electron acceptor portion of the molecule with successively stronger acceptor groups, the scientists demonstrated a material with the best recorded third-order nonlinear response. But it was not all gifts and glory. "We had to move the peak of the absorption to a longer wavelength. So, yes, we increased the nonlinearity, but you can't use it at the visible," said Stegeman, whose research is mainly concerned with wavelengths common to telecommunications networks. Also, the substituted acceptors made the material unstable, so that it required special handling and storage. "The important thing is a root direction," he said. A commercially viable third-order nonlinear material will have efficiencies 10 to 100 times that of the altered b-carotene molecule, Stegeman said. "Now people can go to another family of materials and get improvements with more stability," he said.