Computer Modeling Assists in Predicting Desirable Adhesive Bonding Formulations

Objectives: The long-term objective is to develop adhesive bonding formulations to minimize “recurrent caries” at composite restoration margins. Many contemporary formulations contain cross-linking monomers held together with hydrolyzable ester (-C(=O)O-C-) groups. Some contain components with solubility parameters that differ excessively from one another and with their intended substrates. Our current objectives are to develop bonding formulations containing cross-linking monomers with polar surface-binding “ligand” groups, all held together with hydrolytically stable ether (-C-O-C-) linkages. Other components must also have compatible solubility parameters permitting uniform penetration and interactions with hydrated dentin and enamel. On the basis of these observations, the hypothesis was that three-dimensional modeling and computation of physical-chemical characteristics of candidate adhesives and substrate components would identify improved formulations. Methods: Computer-Assisted Chemistry software was used to develop 3-D models and obtain Log P values of adhesive-bonding formulation components and portions of simulated dentin. These models, with use of the “SMILES” algorithm, were also used to calculate and compare Hoy Solubility Parameters to facilitate synthesis of formulation compounds to improve adhesives for dental and biomedical applications. Results: The 3-D modeling and attendant calculations indicated that the tooth substrates of interest are very hydrophilic, having Log P values below 1.0 and high corresponding Hoy Solubility Parameter values. The modeled active ingredients of currently used bonding formulations did not have similar values for these properties. The hypothesis was supported in that reagent proportions were indicated for synthesizing hydrophilic monomers for use in formulations to penetrate hydrated tooth surfaces and to form densely cross-linked, hydrolytically stable adhesive polymers. Conclusions: The values generated through modeling showed differences in Log P and solubility parameters, which could be used to compare the effects of structural differences, thereby suggesting reagent ratios for use in syntheses of formulation components. This work was supported by NIDCR DE05129, NIST, and the ADA/ADAF.