One dimensional (1D) structures, or nanowires, are expected to play a role in future integrated circuits as both devices and interconnects. One of the most successful approaches for producing nanowires is based on the vapor -liquid-solid (VLS) growth process described nearly 40 years ago by Wagner and Ellis. Several researchers have used adaptations of this method to generate nanowires. One inherent problem with these approaches however, is the formation of entangled meshes of nanowires. While some researchers have made significant progress in manipulating these nanowire meshes into useful configurations for potential electronic devices, other researchers have focused on forming nanowires in predefined architectures to allow easier processing and integration of the nanostructures into functioning devices.

Encapsulation of nanowires within an ordered template offers the possibility of manipulating nanowires into useful configurations and allows their aspect ratios and, hence, their physical properties to be tailored. Polycarbonate track etched membranes, orientated steps at single crystal surfaces and nanochannel array glasses have previously been used as templates for nanowires. These inclusion processes are similar to growth within lithographically defined regions utilized in current device fabrication. While these templating methods are useful, the formation of ordered arrays of nanoscale channels within these templates is difficult and the channel dimensions are often too large to exhibit useful quantum confinement effects. Anodic aluminum oxide and mesoporous materials, however, have cylindrical pore geometry and monodisperse diameters arranged in uniform hexagonal arrays with diameters between 10 – 100 and 1 – 20 nm in diameter, respectively. These porous materials have been exploited as templates for nanomaterials synthesised from chemical vapor deposition (CVD), electrodeposition, and incipient wetness techniques but often have problems with complete pore inclusion due to pore plugging where the surface tension of the liquid solvent prevents precursor penetration into the pores. CVD approaches are less prone to pore plugging but can undergo capillary condensation resulting in liquid phases within the pores. The high-diffusivity, high precursor solubility, and reduced surface tension of supercritical fluids (SCFs), however, result in rapid nucleation and growth of the nanowires within the pores reducing the reaction time for pore inclusion by at least an order of magnitude compared to CVD. Additionally, SCFs cannot be condensed to a liquid phase reducing the problems of pore plugging and incomplete inclusion seen with CVD, electrodeposition, and incipient wetness techniques.