Study of the thickness of adsorbed water layers by atomic force microscopy.

The effect of ambient humidity and adsorbed water can be of critical importance in the processing of fine powders in air. Adsorbed water layers can influence the adhesive properties of the powder and may lead to difficulties in processing and handling. It has been shown, in the current work, that in ambient conditions the interaction between two solid surfaces is dominated by the force arising from the presence of adsorbed water layers. In the current work an atomic force microscopy technique has been developed to determine the separation distance at which two solid surfaces, i.e. the AFM cantilever tip and the sample surface, 'jump' into contact. From the separation distance the thickness of the adsorbed water layers on the cantilever tip and sample surface can be determined based on an analytical method originally developed by Forcada (1993), which considers the interacting forces which cause the 'jump' to contact. The adsorbed layer thickness, as a function of relative humidity, has been determined for silicon wafer, using the AFM technique. This localized adsorption isotherm has been compared with those published by other investigators, who have used different measurement techniques on large sample areas. The adsorption isotherm determined using the AFM technique reports adsorbed layer thicknesses that are significantly larger than those measured by other investigators. Adsorption isotherms have been determined, using the AFM technique, for three forms of a-lactose monohydrate, which is widely used throughout the pharmaceutical industry. It was observed that each form of lactose gave a different level of water adsorption. Laboratory grown crystals exhibited the greatest levels of moisture adsorption, while commercially produced milled a-lactose displayed lower levels of water adsorption. Differences in surface roughness may possibly explain the variation. For rough samples the probe tip makes contact with surface asperities and therefore does not detect the presence of condensed water in the surface valleys. It is also possible that surface contamination and process history may have influenced the levels of water adsorption but these are complex effects to quantify. Water adsorption isotherms were measured on bulk samples of classified lactose, to attempt to validate the AFM technique, developed in the current study. It can be observed that the adsorbed layer thicknesses determined using AFM are significantly larger than those inferred from bulk measurements. The AEM method provides layer thickness values which are approximately four times larger than what are believed to be the true values. This is seen for tests on silicon, aluminium and lactose surfaces. This discrepancy is thought to arise from the presence of the probe tip on the surface causing a local spatial inhomogeneity. The associated increase in surface potential will promote nucleation of water molecules, leading ultimately to capillary condensation