To establish the relationship of resin composition and resin hydrophilicity (indicated by solubility parameters and log P ) to water sorption (WS), solubility, and degree of double bond conversion (DC) of resin mixtures designed for adhesive restoratives by varying the concentration of pyromellitic glycerol dimethacrylate (PMGDM) and various co-monomers.
Sixteen resin mixtures were prepared with (30–70) mass fraction % PMGDM. At given PMGDM concentrations there were up to five compositions with increasing log P . Polymer disks (13 mm × 0.7 mm) were exposed to 96% relative humidity (RH) to determine water sorption in humid atmosphere (WSH) and subsequently immersed in water for immersion water sorption (WSI) and solubility. DC was assessed by near infrared spectroscopy.
WSI was somewhat higher than WSH, which ranged from (2.1 to 5.3) mass fraction %. Both data were positively correlated to PMGDM concentrations [Pearson correlation, p < 0.02; R 2 = 0.74, 0.73 (WSI)] and solubility ( R 2 = 0.64), but not to log P . When grouped by structural similarities, i.e., base resins with bisphenol A core (Group B), Group O containing diluent monomers, or Group U containing urethane dimethacrylate, WS within each group was inversely correlated to log P with R 2 = 0.98, 0.81, 0.95, and WS/solubility correlation improved with R 2 = 0.88, 0.92 and 0.75, respectively. Solubility ranging from 0.3% to 2.3% was inversely related to DC ( r = −0.872). Conversion ranging from 41% to 81% was lower for resins with high base monomer concentrations and highest in mixtures with UDMA.
Log P was a good predictor of WS after grouping the resins according to functional, compositional and structural similarities. WS and conversion were reasonably well predicted from Hoy’s solubility parameters and other physical resin properties.
Earlier work in our group on remineralizing pulp capping and basing or lining materials has led to the development of resin-based cements containing tetracalcium and dicalcium phosphates and carboxylated monomers . The resin matrix of these cements consisted of the adhesive monomer pyromellitic glycerol dimethacrylate (PMGDM) and ethoxylated bisphenol A dimethacrylate (EBPADMA). Due to calcium and phosphate ions, which can diffuse out of the cement into mineral-deficient tooth tissue, several in vitro and in vivo studies have demonstrated that the resin-based Ca-PO 4 cements (RCPC) are capable of remineralizing mineral-deficient dentin . The RCPC has good biocompatibility , and adheres to dentin (about 5 MPa) due to the presence of PMGDM . In an in vivo pulp capping study on healthy dog pulps the formation of dentin bridges demonstrated good response of the pulps to the cement. RCPC has also been used as a pulp capping material on healthy and inflamed ferret pulps. The results on inflamed ferret teeth were particularly encouraging, as the treatment of the inflamed pulps with a commercial light-curing pulp capping agent was less than satisfactory .
This investigation is part of a project to develop materials for improved treatment and prevention of dental caries in Atraumatic Restorative Treatments (ART). This requires restoratives with adhesive properties conveyed through the adhesive monomer PMGDM. Ultimately, the formulations will also contain Ca-PO 4 fillers to provide ion release for the remineralization of the targeted tissues. The main requirements for such materials are high strength and high release of calcium and phosphate ions. Both properties are affected by water sorption and the degree of conversion. Here, it was the objective to establish the relationship of resin composition and resin hydrophilicity (indicated by solubility parameters and log P ) with water sorption (WS), solubility, and degree of double bond conversion (DC) by formulating an array of resin mixtures from varying concentrations of PMGDM and different amounts of the base monomers EBPADMA, bisphenol A glycidyl methacrylate (Bis-GMA), a urethane dimethacrylate (UDMA) monomer, and the two diluent monomers benzyl methacrylate (BMA) and triethyleneglycol dimethacrylate (TEGDMA). Subsequent studies (to be presented in future publications) will evaluate the effects of resin composition on ion release and mechanical properties of composites made from these resins and Ca-PO 4 fillers. Ultimately selected composites will be tested for their ability to remineralize artificial and natural caries lesions.
Materials and methods
Sixteen resin compositions spanning a range of hydrophilicities were formulated from various PMGDM concentrations by combining PMGDM with various monomers typically used in dental materials at different concentrations. The monomer structures are depicted in Fig. 1 , and the compositions in mass fraction % are listed in Table 1 . The octanol/water partition coefficients (log P ) as a measure of hydrophilicity were calculated for each monomer using Chem Draw (CambridgeSoft Corporation, Cambridge, MA, USA) and then assessed for each resin mixture according to the composition. The mixtures were formulated to have in common either identical PMGDM concentrations or log P values ( Fig. 2 ). For any given PMGDM concentration there were up to five compositions with increasing log P . At any given log P value there were one to four compositions with increasing PMGDM concentration. The log P ranged from 3.2 to 5.2, the PMGDM concentrations from 30 to 70 mass fraction %. Each resin was photo-activated with 0.6 mole fraction % camphorquinone and 2.0 mole fraction % of the tertiary amine ethyl 4-(dimethylamino)benzoate. For each composition three polymer disks (13 mm in diameter and 0.7 mm in height) were prepared by photocuring the Mylar-covered resin mixtures in Teflon molds for 1 min per side using a dental light (QHL75 Curing Light, Dentsply Int. Milford, DE) at a distance of about 5 mm from the specimen surface. After removing the disks from the molds, they were dried over Drierite to constancy until the mass changed less than 0.1 mg. The specimens were subjected to one sorption/desorption cycle by exposing them for up to 1 month to a relative humidity (RH) of 96% above a potassium sulfate slurry at 23 °C and then redrying them over Drierite. The water sorption after exposure to humidity (WSH) was determined gravimetrically and calculated as mass fraction %.
|PMGDM||EBPADMA||Bis-GMA||UDMA||BMA||TEGDMA||Log P||HLB||δ t||δ p||δ h|
For the diffusion coefficient D , the slopes M t / M e were plotted against t 1/2 where M t is the mass gain of the specimen at time t , and M e is the mass gain at equilibrium.
The diffusion coefficient was calculated for the initial stages of water sorption when M t / M ∞ ≤ 0.5 from:
M t M e = 2 Dt π l 2 1 / 2
Subsequently, the specimens were immersed in water at 37 °C and measured until equilibrated (up to 12 d). They were then dried back over Drierite until they reached a steady weight. The immersion water sorption (WSI) in mass fraction % was determined gravimetrically from:
WSI % = m 2 − m 3 m 1 × 100
where m 1 = initial conditioned mass prior to humidity exposure, m 2 = mass after immersion and m 3 = reconditioned (redried mass). The solubility was calculated from:
Solubility % = m 1 − m 3 m 1 × 100
The degree of conversion was measured by near infrared spectroscopy (NIR) , which also provides the peak position of the water peak at about 5200 cm −1 and allows conclusion about the hydrophilic/hydrophobic properties of the polymer matrix . Hoy’s total solubility parameter δ t (delta t ) and the polar and hydrogen bonding contributions of the neat resins were estimated with the use of a Computer-Aided Chemistry system (CAChe 6.1.1 WorkSystem Pro including BioMedCAChe, Fujitsu Ltd., Beaverton, OR). The solubility parameters were then determined for each resin mixture .
The solubility parameters of copolymers, δ c , were calculated from:
δ c = ∑ δ i m i
where δ i is the solubility parameter of the homopolymer that corresponds to the monomer i , and m i is the mass fraction of monomer i in the copolymer. Solubility (miscibility) of two monomers is expected if the value of ( δ 1 − δ 2 ) is less than unity. This was the case for all monomer mixtures. Mean water sorption and related data, together with their standard deviations to express the standard uncertainty, are reported. Data were evaluated statistically at α = 0.05 by one-way analysis of variance and all pairwise multiple comparison procedures (Holm–Sidak method). Where appropriate, multiple regression analysis and Pearson product moment correlation coefficients R and the corresponding p values were calculated ( p < 0.05).