The silicon wafer hydrophobized with OTS was immersed into water to observe the surface in-situ by tapping-mode AFM. A large number of nano-size domain images were found on the surface. Their shapes were characterized by the height image procedure of AFM, and the differences of the properties compared to those of the bare surface were analyzed using the phase image procedure and the interaction force curves. All the results consistently implied that the domains represent the nanoscopic bubbles attached on the surface. This was confirmed by the fact that no domain was observed in the case of the surfaces hydrophobized in the AFM fluid cell without exposure to air. The apparent contact angle of the bubbles was much smaller than that expected macroscopically, which was postulated to be the reason bubbles were able to sit stably on the surface.

We report grand canonical Monte Carlo simulations for a Lennard-Jones (LJ) fluid modeled on methane in slit-shaped pores of several materials and pore widths. Three types of pore wall were considered: graphitic carbon (strongly attractive walls), “methane’’ walls (wall attractions equal to those in the adsorbate phase), and hard walls. For each system the change from a fluidlike to a solidlike adsorbed phase was observed, and the shift in freezing or melting temperature from that of the bulk adsorbate material was determined. As well as changes in the overall properties of the adsorbate phase, corresponding changes in the individual adsorbate layers in the pore were studied. In addition hysteresis on heating and cooling was examined. For the graphitic carbon walls the freezing temperature was raised relative to that of the bulk material, the elevation being greater for smaller pore widths; however, no freezing transition was observed for pore widths below about 5.3σ. In addition, the contact layer of adsorbate froze at a temperature higher than the inner layers. For pores with methane walls (walls of LJ molecules having the same density and intermolecular interactions as the adsorbate phase) no shift in freezing temperature occurred, while pores with hard walls showed a decrease in freezing temperature relative to the bulk; in the case of hard walls, the contact layer of adsorbate froze at a lower temperature than the inner layers. Considerable hysteresis was observed in some cases, and the width of the hysteresis loop was sensitive to pore size, being wider for pores in which the adsorbed layers are tightly packed. The results indicate that the direction and magnitude of the shift in freezing temperature in the pore is strongly dependent on the strength of the attractive forces between the adsorbate molecules and the wall, and particularly on the magnitude of this relative to such forces between the adsorbate and a wall composed of the same adsorbate molecules. A simple thermodynamic model based on this idea is proposed, and showed to give a good account of the simulation results for methane in carbons. In the simple systems studied here the confinement causes little change in the solid lattice structure of the bulk material. This is unlikely to be the case for more complex pore geometries, and the analysis of such cases is likely to involve additional structural effects.

The thermal transitions of confined polymers are important for the application of polymers in molecular scale devices and advanced nanotechnology. However, thermal transitions of ultrathin polymer assemblies confined in subnanometre spaces are poorly understood. In this study, we show that incorporation of polyethylene glycol (PEG) into nanochannels of porous coordination polymers (PCPs) enabled observation of thermal transitions of the chain assemblies by differential scanning calorimetry. The pore size and surface functionality of PCPs can be tailored to study the transition behaviour of confined polymers. The transition temperature of PEG in PCPs was determined by manipulating the pore size and the pore–polymer interactions. It is also striking that the transition temperature of the confined PEG decreased as the molecular weight of PEG increased.

To clarify the origin of the long-range attraction between hydrophobic surfaces in water, the interaction between the surfaces silanated by the popular method (type I) and that between the surfaces silanated without exposing to air (type II) were examined using an atomic force microscope (AFM) and their characteristics were compared. The interaction between type I surfaces was long-ranged, and a discontinuous step appeared in the approaching and separating force curves, respectively, whereas the interaction between type II surfaces was short-ranged and no step was found. Once type II surfaces were exposed to air, however, the similar interaction to that for type I surfaces appeared. As for type I surfaces, the force curves depended on the local property of the surface, and the interaction in the first cycle of force measurements differed from those in the later cycles. These findings enabled us to estimate the following mechanism for the long-range attraction. When surfaces are hydrophobized, they are usually exposed to air during the hydrophobizing reaction or in the drying process. Then they are immersed in water to measure the interaction without removing microscopic bubbles on the surfaces completely. These bubbles coalesce before the surfaces contact and generate a strong long-range interaction. Hence, this interaction is not the genuine hydrophobic attraction.

We have studied the formation of stripe patterned films of ordered particle arrays on completely solvophilic substrates by using a self-organization technique. In this method, a substrate immersed in a suspension is withdrawn vertically at a controlled temperature. We have also systematically examined the effects of several experimental parameters. Well-defined stripes spontaneously form at the air-solvent-substrate contact line because of a very dilute suspension in a quasi-static process. The stripe width depends on particle concentration, withdrawal rate, and surface tension, while the stripe spacing depends on the thickness of stripes, surface tension, and type of substrate. A stripe width and the adjacent spacing show a clear correlation, strongly indicating the synchronized formation of a stripe and the next spacing. The evaporation rate does not affect stripe width and spacing but determines the growth rate of stripe patterned films. Based on these results, we propose a new mechanism for stripe formation, which is neither a stick-slip motion of the contact line nor dewetting but a negative feedback of particle concentration caused by a concavely curved shape of the meniscus, demonstrating not only its qualitative but also its quantitative validity.