Surface absorption spectroscopy is used for everything from industrial applications to environmental studies to biotechnology, as well as for basic research of thin-film growth and of light-matter interaction. Typically, the technique is performed by recording the attenuation of a freely propagating laser beam by a surface covered with the substance under study. Alternatively, the substance can be deposited on the surface of a specially designed glass fiber. Compared with freely propagating beams, fibers enable greater control of the beam direction and, when the fiber diameter is varied, the speed as well.Searching for the Ultimate Sensitivity of Surface Absorption Spectroscopy An ultrathin optical fiber, such as the one shown, can enable surface absorption spectroscopy measurements that are much more sensitive than those made with free-beam spectroscopy. Courtesy of Marcus Kaufhold.However, so far no one has investigated the upper limit of the sensitivity of fiber-based surface absorption spectroscopy, according to researchers at Universität Bonn in Germany, who aim to determine that limit. Toward that goal, they made theoretical calculations and performed experiments analyzing spectroscopic measurements of thin films of 3,4,9,10-perylene-tetracarboxylic dianhydride (PTCDA) on ultrathin glass fibers.The principal investigator, Arno Rauschenbeutel, now at Universität Mainz, also in Germany, said that they chose the perylene-based compound for several reasons. For one, it is chemically stable in air, whereas the makeup of other chemicals might be changed by reacting with oxygen. Therefore, it can sublimate and adsorb to the fiber in ambient conditions, making it easy to work with. Additionally, it is a chromophore with a spectrum known from free-beam spectroscopy, but that technique did not characterize it with the desired sensitivity. Finally, the knowledge of the absorption cross section from the compound enabled the researchers to use spectroscopy to quantify how much of the compound they were working with.In their theoretical analysis, they predicted that fiber-based spectroscopy could be performed with 10,000 times more sensitivity than with free-beam spectroscopy if they used a fiber with a length on the order of millimeters and a submicron radius, smaller than the wavelength of the guided light.In their experiments, they thinned the center of a standard optical fiber by stretching the fiber with a homemade fiber-pulling rig while heating the center over a flame. Using this process, they produced a fiber with an ultrathin waist that was 500 nm in diameter and 3 mm long, which they used for their experiment. The process resulted in a fiber that efficiently delivered light, the researchers said. As proof, they noted that it could transmit 97 percent of light with an 850-nm wavelength.After heating PTCDA molecules to a gas, the material adsorbed to the waist and began forming a thin film, much less than a single molecular layer thick. For spectroscopy, white light from a common tungsten lamp passed through the fiber to a CCD spectrograph from Ocean Optics Inc. of Dunedin, Fla.They achieved a signal-to-noise ratio 100 times greater than that of the free-beam spectra of the same compound. They monitored the formation of thin films on the fiber with a temporal resolution of 1 s, whereas free-beam spectroscopy could monitor the process with timescales of only minutes.Rauschenbeutel noted that they achieved the high sensitivity with a relatively inexpensive setup, costing somewhere on the order of €4200, excluding the fiber-pulling rig. He said that they could reach the theoretical sensitivity using a white-light source and a spectrometer with lower noise.He added that these experiments will provide a foundation for using the fiber to manipulate atoms and light at the quantum mechanical level. Light traveling through the fiber could influence the position of atoms; or, in a separate line of thought, atoms could act as a light switch, turning light on or off. Controlling light and matter in this way could enable quantum communication and computation, which promise to become next-generation ways to deliver and to process information, respectively.Optics Express, Sept. 17, 2007, pp. 11952-11958.