Objectives: The aim of this study was to test the null hypothesis that the tensile properties of demineralized dentin are not influenced by the hydrogen bonding ability of anhydrous polar solvents.
Methods: Dentin disks 0.5mm thick were prepared from mid-coronal dentin of extracted, unerupted, human third molars. 'I' beam and hour-glass shaped specimens were prepared from the disks, the ends protected with nail varnish and the central regions completely demineralized in 0.5M EDTA for 5 days. Ultimate tensile stress (UTS) and low-strain apparent modulus of elasticity (E) were determined with the specimens immersed for 60 min in water, methanol, HEMA, acetone or air prior to testing in those same media. Apparent moduli of elasticity were measured on the same specimens in a repeated measures experimental design. The results were analyzed with a one-way ANOVA on ranks, followed by Dunn's test at alpha=0.05. Regression analysis examined the relationship between UTS or E and Hansen's solubility parameter for hydrogen bonding (delta(h)) of each solvent.
Results: The UTS of demineralized dentin in water, methanol, HEMA, acetone and air was 18(7), 29(7), 31(6), 41(13) and 146(27)MPa, x(SD), n=10. Low-strain E for the same media were 11(7), 43(12), 79(21), 132(31) and 253(115)MPa. Regression analysis of UTS vs delta(h) revealed a significant (p<0.0005, r=-0.69, R(2)=0.48) inverse, exponential relationship. A similar inverse relationship was obtained between low-strain E vs delta(h) (p<0.0001, r=-0.93, R(2)=0.86).
Significance: The tensile properties of demineralized dentin are dependent on the hydrogen bonding ability of polar solvents (delta(h)). Solvents with low delta(h) values may permit new interpeptide H-bonding in collagen that increases its tensile properties. Solvents with high delta(h) values prevent the development of these new interpeptide H-bonds.