As electronics continue to shrink, they're constantly pushing up against the limits of our ability to craft increasingly tiny features. Processors rely on features etched with extremely high-energy light, and disk drives store information in ever-smaller clusters of atoms. As these features shrink, electrical, magnetic, and even quantum interference begin to dominate, and it becomes ever more difficult to maintain and detect signals such as the state of a memory bit. To avoid these issues entirely, scientists have started to explore the possibility of storing information in the chemical structure of single molecules. A team of European researchers reported a new approach to single-molecule storage that may bring these devices closer to stepping out of the lab.

Molecular memory will require chemicals that can switch back and forth between two stable states, much the way clusters of atoms switch magnetic states on the surface of disk drives. It's relatively easy to find molecules that can behave the same way. So far, however, most of these molecular switches involve structural changes: large parts of the molecule move relative to each other when changing states. This can work well in lab settings, but it isn't ideal in the real world, where it may not be compatible with a stable and reliable material that's easy to manufacture. The approach described in the new research relies on a molecule that's physically flat and changes states by shifting the location of hydrogen atoms without undergoing any structural changes. Even better, the memory states can be changed and read through the same technique used in electronics today: changes in electrical conduction.


The new work is based on a compound called naphthalocyanine, a cross-shaped molecule consisting of a number of interlocking ringed structures. At the center of the cross, four nitrogen atoms face inward; two of those nitrogens, located opposite each other, are bonded to hydrogen atoms. The key fact is that it doesn't matter which two. There are two equally stable conformations of the molecule, termed tautomers, that differ only in the location of these hydrogen atoms.

The authors layered these molecules on an insulating surface and chilled them to five degrees above absolute zero. They found that a scanning-tunneling microscope could readily detect the axis of the molecule that included the two hydrogens. By manipulating the current tunneling out of the microscope's tip, however, they were able to induce the hydrogens to swap locations. Because of the interlinked chemical structure of naphthalocyanine, energy pumped anywhere into the rings was able to induce this "tautomerization." In fact, putting energy into the end of one of the molecule's arms was the most efficient way of moving the hydrogens to a different location. Overall, the researchers claim that they can accurately set the state of these molecular bits 90 percent of the time.

The team was not content to stop at a single molecular bit, however. They used the tip of the tunneling microscope to push several naphthalocyanine molecules close enough that the electrons in their ringed structures formed linked orbitals. Depending on where current is injected in the rings, different members of the structure can be selectively switched. They suggest that similar arrangements might also either allow the coordinated switching of a number of atomic bits, or enable the state of one bit to influence the response of its neighbors.

Clearly, this approach is not ready for use on the desktop. It operates just a shade above absolute zero, requires a scanning-tunneling microscope, and only sets bits with 90 percent accuracy. Still, as the authors suggest, the basic approach—one where the molecule holding the bit remains largely unperturbed by changes in its value—is far more likely to produce usable technology than many of the approaches that have been described previously.