The way how liquids infiltrate nanoporous and microporous templates determines the morphology of the one-dimensional nanostructures and microstructures to a large extent. Wetting pores by liquids consisting of molecules with low molecular weight starts with the formation of a wetting layer on their walls, similar to the formation of precursor films on smooth substrates (P. G. de Gennes, Rev. Mod. Phys. 1985, 57, 827). However, curved liquid films are susceptible to instabilities (Lord Rayleigh, Phil. Mag. Série 5 1892, 34, 177). Their occurrence leads to the formation of menisci (“snap-off”), and starting from the menisci to a complete filling of the pore volume, as more liquid moves into the pore (M. G. Bernadiner, Transport in Porous Media 1998, 30, 251). Eventually, this results in the complete filling of the pore volume (Figure 1). If the liquid being infiltrated is a melt containing sufficiently mobile macromolecules, and if the pore diameter is larger than the characteristic dimensions of the macromolecules, a mesoscopic wetting layer will rapidly form on the pore walls. The polymeric film with a typical thickness of a few tens of nm is kinetically stable, and tubes can easily be obtained by solidifying the polymer (Figure 2). As this mechanism involves the formation of a precursor film, it can be referred to as “precursor wetting”. A prerequisite for “precursor wetting” is a sufficiently high surface energy of the pore walls.
Figure 1. Infiltration of porous templates by “precursor wetting”. A mesoscopic precursor film forms rapidly on the pore walls. Instabilities occur and grow, as more liquid moves into the pores, eventually leading to the complete filling of the pore volume.
Figure 2. Infiltration of polymers by “precursor wetting”. A mesoscopic precursor film forms rapidly on the pore walls, which is kinetically stable. Solidification of the polymer yields nanotubes.
If templates are wetted with polymers at temperatures slightly above their solidification temperatures (M. Zhang, P. Dobriyal, J.-T. Chen, T. P. Russell, J. Olmo, A. Merry, Nano Lett. 2006, 5, 1075-1079) or by microphase-separated block copolymer melts (H. Q. Xiang, K. Shin, T. Kim, S. I. Moon, T. J. McCarthy, T. P. Russell, Macromolecules 2004, 37, 5660), a solid liquid thread preceded by a meniscus moves into the pore (Figure 3). Apparently, the rapid formation of a precursor film is suppressed, when individual macromolecules cannot easily be removed from the bulk. The same mechanism governs infiltration if the initially polar pore walls are converted into a non-polar surface with low surface energy, for example by silanization. Then, there is no more driving force for the rapid formation of a precursor film. As this mechanismus is driven by classical capillarity, it may be referred to as “capillary wetting”
Figure 3. Infiltration of polymers by “capillary wetting”. No precursor film forms because either the mobility of the polymer molecules or the surface energy of the pore walls is reduced. Instead, the liquid polymer infiltrates the pores as a solid thread preceded by a meniscus.
As example in case, Fig. 4 shows nanofibers and microfibers consisting of the commodity polymer polystyrene. “Precursor wetting” yields tubular nanostructures, as the polystyrene nanotube protruding from a self-ordered porous alumina template seen in Fig. 4a. Decreasing the surface energy of the pore walls in macroporous silicon, for example by silanization, yields solid microrods formed according to capillary wetting (Fig. 4b).
Figure 4. Polystyrene fibers prepared by template wetting. a) Broken polystyrene nanotube protruding from an ordered porous alumina template prepared by “precursor wetting”. b) Released polystyrene microrods obtained by “capillary wetting”, using a template with silanized pore walls.
An example, where the wetting of the pore walls is driven by chemical spreading, is the formation of lithium niobate microtubes inside the pores of macroporous silicon. This material shows pronounced non-linear optical properties and is an excellent host for trivalent erbium, whose emission maximum coincides with that of the transmittance maximum of silica-based photonic building blocks. Figure 5a shows a scanning electron microscopy image of released lithium niobate microtubes, Fig. 5b a transmission electron microscopy image of a segment of a lithium niobate microtube, and Fig. 5c the corresponding selected area electron diffraction pattern. It is obvious that the tube segment is single-crystalline. The lithium niobate microtubes are doped with 0.3 mol-% Er and show the characteristic photoluminescence of dispersed Er3+.
Figure 5. Erbium-doped lithium niobate microtubes. a) Scanning electron microscopy image of released erbium-doped lithium niobate microtubes; b) transmission electron microscopy image of a segment of a single erbium-doped lithium niobate microtube; and c) corresponding selected area electron diffraction pattern, evidencing the single-crystalline nature of the tube segment seen in b).