Epitaxial self-assembly of block copolymers on lithographically defined nanopatterned substrates

Parallel processes for patterning densely packed nanometre-scale structures are critical for many diverse areas of nanotechnology. Thin films of diblock copolymers can self-assemble into ordered periodic structures at the molecular scale (approximately 5 to 50 nm), and have been used as templates to fabricate quantum dots, nanowires, magnetic storage media, nanopores and silicon capacitors. Unfortunately, perfect periodic domain ordering can only be achieved over micrometre-scale areas at best and defects exist at the edges of grain boundaries. These limitations preclude the use of block-copolymer lithography for many advanced applications. Graphoepitaxy, in-plane electric fields, temperature gradients, and directional solidification have also been demonstrated to induce orientation or long-range order with varying degrees of success. Here we demonstrate the integration of thin films of block copolymer with advanced lithographic techniques to induce epitaxial self-assembly of domains. The resulting patterns are defect-free, are oriented and registered with the underlying substrate and can be created over arbitrarily large areas. These structures are determined by the size and quality of the lithographically defined surface pattern rather than by the inherent limitations of the self-assembly process. Our results illustrate how hybrid strategies to nanofabrication allow for molecular level control in existing manufacturing processes.     

Mixed parvocellular and magnocellular geniculate signals in visual area V4.

Visual information from the retina is transmitted to the cerebral cortex by way of the lateral geniculate nucleus (LGN) in the thalamus. In primates, most of the retinal ganglion cells that project to the LGN belong to one of two classes, P and M, whose axons terminate in the parvocellular or magnocellular subdivisions of the LGN. These cell classes give rise to two channels that have been distinguished anatomically, physiologically and behaviourally. The visual cortex also can be subdivided into two pathways, one specialized for motion processing and the other for colour and form information. Several lines of indirect evidence have suggested a close correspondence between the subcortical and cortical pathways, such that the M channel provides input to the motion pathway and the P channel drives the colour/form pathway. This hypothesis was tested directly by selectively inactivating either the magnocellular or parvocellular subdivision of the LGN and recording the effects on visual responses in the cortex. We have previously reported that, in accordance with the hypothesis, responses in the motion pathway in the cortex depend primarily on magnocellular LGN. We now report that in the colour/form pathway, visual responses depend on both P and M input. These results argue against a simple correspondence between the subcortical and cortical pathways.