Mechanical properties of in situ demineralised human enamel measured by AFM nanoindentation

Abstract

Diet-induced demineralisation is one of the key factors in surface changes of tooth enamel, with soft drinks being a significant etiological agent. The first step in this dissolution process is characterised by a change in the mechanical properties of the enamel and a roughening of the surface. The objective of this pilot study was to measure early stages of in situ induced hardness changes of polished human enamel surfaces with high accuracy using a nanoindenter attached to an atomic force microscope (AFM). Human unerupted third molars were cleaned, sterilised with sodium hypochlorite, sectioned and embedded in epoxy resin. The outer enamel surface was polished and the samples partly covered with a tape, allowing a 2-mm-wide zone to be exposed to the oral environment. Samples were fitted in an intra-oral appliance, which was worn from 9 a.m. to 5 p.m. for one day. During this time the volunteer sipped 250 ml of a drink over 10 min periods at 9.00, 11.00, 13.00 and 15.00 h. Three different drinks, mineral water, orange juice and the prototype of a blackcurrant drink with low demineralisation potential were used in this study. At the end of the experiment the samples were detached from the appliance, the tape removed and the surfaces chemically cleaned. The surface hardness and reduced Young's modulus of the exposed and unexposed areas of each sample were determined. In addition, high resolution topographical AFM images were obtained. This study shows that by determining the hardness and reduced Young's modulus, the difference in demineralisation caused by the drinks can be detected and quantified before statistically significant changes in surface topography could be observed with the AFM. The maximum decrease in surface hardness and Young's modulus occurred in the samples exposed to orange juice, followed by those exposed to the blackcurrant drink, while exposure to water led to the same values as unexposed areas. A one-way ANOVA showed a statistically significant difference between the changes in hardness for the drinks at a 95% confidence level ( p=0.0000), while a Kruskal–Wallis test proved a statistically significant difference between the changes in the reduced Young's modulus at a 95% confidence level ( p=0.0000). Thus, it was possible to detect differences in demineralisation potential in an in situ study at a very early stage. Further investigations with greater subject numbers and a larger quantity of samples are necessary to fully evaluate the potential of this method.