Roughness of mesoporous silica surfaces deduced from adsorption measurements

Abstract

The complete understanding of the properties of a material goes in close relationship with a correct description of their structure. This is particularly true for mesoporous silica that exhibits different shapes and sizes of the porous framework as well as different degrees of crosslinking of the silica network itself. These variations can affect drastically the specific properties of these materials as well as their hydrothermal resistance. Hydrothermal resistance and adsorption properties are governed by the surface structure, but the later parameter is not frequently studied because it can be difficult to characterize. Therefore a simple parameter such as surface roughness could be helpful in relating the properties of materials with their surface structure. The direct determination of surface roughness is impossible, but it has been demonstrated that a single mathematical parameter, the surface fractal dimension Ds, can be related within a certain limit with surface roughness. Here we demonstrate using nitrogen adsorption data that a correlation can be drawn between the values of the surface fractal dimension Ds used as a gauge of the surface roughness and the structural evolution of a series of mesoporous silicas. Among the different adsorption models for obtaining Ds that are available nowadays, we selected the one proposed by Avnir and Jaroniec. We demonstrate in this report that there exists a real correlation between the evolution of Ds, when applied to a series of samples prepared under different synthesis conditions, with the evolution of their structure and properties. By example, this study allowed us to link the range of hydrothermal resistance of different mesoporous silicas (e.g. the HMS, MSU, MCM, SBA families of silicas) with their surface roughness. Moreover, we found that HMS silica, a very resistant material toward hydrothermal treatment, exhibits actually the lowest surface roughness. We could demonstrate too how different synthesis parameters such as temperature, can modify this surface roughness, hence the final properties of the mesoporous silica. Finally, we show that this method can even be applied to water ellipsoporometry and help to follow the crystallization of thin layers of metal oxide. Bearing in mind that Ds is a parameter that can readily be determined using the existing software – based on different models – associated with most commercial adsorption equipment, we urge the broader use of the Ds parameter for the structural characterization of surface roughness in mesoporous materials and metal oxide thin layers.