The first man-made single-molecule machine has proved it can do mechanical work, according to its German builders. Hermann Gaub, a researcher with the University of Munich's Nanoscience Center and a member of the project team that developed the photopolymer/silicon hybrid machine, has reported that the heart of the system is a single photosensitive polymer that contracts or expands upon illumination with 365- or 420-nm light. The polymer "works" by driving a silicon cantilever several tens of microns in size (see figure).In this hybrid polymer/silicon single-molecule machine, an external force acting along the polymer backbone causes the polymer to contract, which in turn drives movement of the cantilever.The photoactive system in the machine is a polymer of azobenzene units that lengthens and contracts when optically switched between trans (long) and cis (short) configurations. The scientists studied the individual polymer strands using single-molecule force spectroscopy in combination with optical excitation. To ensure stable attachment, they covalently coupled the polymer end groups to both the tip of the atomic force microscope and a supporting glass slide.They also designed the system for optical excitation by total internal reflection to avoid thermomechanical disturbances related to the cantilever absorbing any optical stimulus. This involved directly focusing light from a high-pressure mercury lamp onto the slide edge at an aperture below the critical angle of total internal reflection, which caused the slide to act as a waveguide.To demonstrate optomechanical energy conversion, Gaub and his colleagues stretched a single polymer strand with a bias force. The polymer contracted upon optical excitation, in the process doing mechanical work. With the external force reduced and the polymer optically switched back to its original position, the operating cycle returned to its starting point. In one experiment with the machine, for state I of the operating cycle, the scientists applied five 420-nm pulses and a force of 80 pN to the polymer. They expanded the strand mechanically to a force of 200 pN for state II. Five 365-nm pulses signaled the polymer to contract against the external force and brought it to state III. They reduced the applied force on the polymer to 85 pN (state IV) and completed the cycle with five more 420-nm pulses, which expanded the molecule to its original length.Gaub said that the work output was the mechanical energy related to the contraction of the polymer band against the external load; i.e., the force applied by the cantilever bar. The work took place as the polymer transitioned from state II to III. He estimated that this was approximately 4.5 x 10–20 J for the cycle detailed."We designed this proof-of-principle experiment with the goal of better understanding the basics of single-molecule optomechanics," Gaub said. "We anticipate that several more years of research will be necessary to develop the necessary understanding of the fundamental principles involved and expect that it will be some time before the concept sees use in commercial nanotechnology applications. As a molecular actuator, however, such a system could operate nanometer-sized mechanical units like valves. Connected to other mechanical devices, such molecular machines could act as pumps or molecular motors. Interconnected in networks, such devices could possibly solve even more complex tasks."Near-term goals for the researchers include improving understanding of the crosstalk between optical and mechanical excitation in the isomerization process. They also are exploring ways to increase the chemical stability and raise the conversion efficiency of the single-molecule machine. Gaub said that the overall efficiency of optomechanical switching for one cycle of the test setup was only on the order of 10–3."Success with this proof-of-concept experiment has also opened the door to a whole new set of topics for exploration," he said. For example, the maximum force that can be generated in biological motors is on the level of just a few piconewtons, but with the light-activated single-molecule machine, the limit (before cis-tran switching is no longer observable) is more on the order of 500 pN. Raising this limit could further expand potential applications for the single-molecule machine.