“We believe that self-assembling materials used in conjunction with the most advanced exposure tools may enable extension of current manufacturing practices to dimensions of 10 nm and less,” Paul Nealey of the University of Wisconsin told nanotechweb.org.
The technique uses a blend containing a diblock copolymer and additional amounts of its two constituent homopolymers. Patterning the polymer mixture onto a chemically altered region of a silicon substrate produced very fine feature sizes.
“Diblock copolymers consist of two polymer chains connected at one end that tend to form ordered structures, including spheres, cylinders and lamellae,” said Nealey. “Previously these self-assembling materials have been used to fabricate periodic arrays at the nanoscale. [In our work] we demonstrate that by directing the assembly of blends of block copolymers and homopolymers on chemically nanopatterned substrates, it is possible to pattern perfect, registered, non-regular shaped structures facilitated by the local redistribution of homopolymer across the patterns.”
Nealey and colleagues used a mix containing 60 wt% symmetric polystyrene-block-poly(methyl methacrylate), 20 wt% polystyrene and 20 wt% poly(methyl methacrylate). They applied the mix to a silicon substrate coated with hydroxy-terminated polystyrene that had been chemically patterned using advanced lithography and an oxygen plasma. This patterning rendered regions of the substrate hydrophilic.
The polystyrene domain of the polymer blend preferentially wet the unpatterned areas of the substrate, while the poly(methyl methacrylate) domain preferred the chemically modified regions. Annealing the polymer blend at 193°C for a week enabled it to reach a stable morphology.
“The extension of block copolymer lithography to pattern features more complex than simple periodic arrays creates opportunities for the use of these non-traditional imaging materials for the production of nanoelectronic devices,” said Nealey.
According to Nealey, there is still significant research needed to optimize blend compositions, materials and interfacial interactions, understand the fundamental physics of polymer blends in thin films, and investigate the achievable surface density of bends or other non-regular geometries. “It is also possible that the design of circuit elements could be adapted to be more amenable for manufacturing using self-assembling materials,” he said. “Single chain in mean-field simulations may, some day, allow for the prediction of the optimal block copolymer or blend to achieve device or pattern-specific self-assembled templates.”
The researchers reported their work in Science.