Block copolymers — molecules that spontaneously self-assemble into useful shapes — are a promising alternative to photolithography techniques for manufacturing nanoscale circuits. Researchers from the Massachusetts Institute of Technology have described what they say is the first technique for stacking layers of block-copolymer wires such that the wires in one layer naturally orient themselves perpendicularly to those in the layer below. The ability to easily produce such "mesh structures" could make self-assembly a much more practical way to manufacture memory, optical chips and even future generations of computer processors, the researchers said. In a block copolymer, the constituent polymers are chosen so that they're chemically incompatible with each other. It's their attempts to push away from each other — both within a single polymer chain and within a polymer film — that causes them to self-organize. In the MIT team’s case, one of the constituent polymers was carbon-based and the other, silicon-based. In an effort to escape the carbon-based polymer, the silicon-based polymers folded in on themselves, forming cylinders with loops of silicon-based polymer on the inside and the other polymer bristling on the outside. When the cylinders were exposed to an oxygen plasma, the carbon-based polymer burned away and the silicon oxidized, leaving glass-like cylinders attached to a base. To assemble a second layer of cylinders, the researchers repeated the process, using copolymers with slightly different chain lengths. The cylinders in the new layer naturally oriented themselves perpendicularly to those in the first. Chemically treating the surface on which the first group of cylinders was formed caused them to line up in parallel rows. In that case, the second layer of cylinders also formed parallel rows, perpendicular to those in the first. The researchers said if the cylinders in the bottom layer were allowed to form haphazardly, snaking out into elaborate, looping patterns, the cylinders in the second layer maintain their relative orientation. These create their own elaborate, but perpendicular, patterns. The orderly mesh structure is the one that has the most obvious applications, but the disorderly one is the more impressive technical feat. This is the one that most excites materials scientists, said graduate student researcher Sam Nicaise. Glass-like wires are not directly useful for electronic applications, but it might be possible to seed them with other types of molecules, which would make them electronically active, or to use them as a template for depositing other materials, the researcher said. They hope to reproduce their results with more functional polymers. To that end, they had to theoretically characterize the process that yielded their results. "We use[d] computer simulations to understand the key parameters controlling the polymer orientation," said graduate student researcher Karim Gadelrab. What they found was that the geometry of the cylinders in the bottom layer limited the possible orientations of the cylinders in the upper layer. If the walls of the lower cylinders were too steep to permit the upper cylinders from fitting in comfortably, the upper cylinders tried to find a different orientation. The researcher said it was also important that the upper and lower layers had only weak chemical interactions. Otherwise, the upper cylinders tried to stack themselves on top of the lower ones like logs on a pile. Both of these properties — cylinder geometry and chemical interaction — can be predicted from the physics of polymer molecules, according to the researchers, meaning it should be possible to identify other polymers that will exhibit the same behavior. The research was published in Nature Communications (doi: 10.1038/ncomms10518).