Abstract
Objectives
The objective of this study was to evaluate ternary methacrylate–thiol–ene systems, with varying thiol–ene content and thiol:ene stoichiometry, as dental restorative resin materials. It was hypothesized that an off-stoichiometric thiol–ene component would enhance interactions between the methacrylate and thiol–ene processes to reduce shrinkage stress while maintaining equivalent mechanical properties.
Methods
Polymerization kinetics and functional group conversions were determined by Fourier transform infrared spectroscopy (FTIR). Cured resin mechanical properties were evaluated using a three-point flexural test, carried out with a hydraulic universal test system. Polymerization shrinkage stress was measured with a tensometer coupled with simultaneous real-time conversion monitoring.
Results
The incorporation of thiol–ene mixtures as reactive diluents into conventional dimethacrylate resins previously was shown to combine synergistically advantageous methacrylate mechanical properties with the improved polymerization kinetics and reduced shrinkage stress of thiol–ene systems. In these systems, due to thiol consumption resultant from both the thiol–ene reaction and chain transfer involving the methacrylate polymerization, the optimum thiol:ene stoichiometry deviates from the traditional 1:1 ratio. Increasing the thiol:ene stoichiometry up to 3:1 results in systems with equivalent flexural modulus, 6–20% reduced flexural strength, and 5–33% reduced shrinkage stress relative to 1:1 stoichiometric thiol:ene systems.
Significance
Due to their improved overall functional group conversion, and shrinkage stress reduction while maintaining equivalent flexural modulus, methacrylate–thiol–ene resins, particularly those with excess thiol, beyond the conventional 1:1 thiol:ene molar ratio, yield superior dental restorative materials compared with purely dimethacrylate resins.
1
Introduction
The vast majority of resin-based composite (RBC) dental restoratives are comprised of dimethacrylate resins. These formulations are functional and have resulted in significant advancements in the field of dentistry. However, there are a number of properties of the resin chemistry that, if improved upon, would increase the performance, longevity and biocompatibility of composite restorations . The most significant shortcomings of methacrylate-based resins are high volumetric shrinkage , high polymerization stress and low functional group conversion . Additionally, the residual monomer left in the restoration after curing is extractable and has the potential to leach from the restoration into the body, with undesirable consequences . As one example, there is concern that residual monomers may cause allergic reactions and sensitization in patients . Additionally, there is reason to believe that release of the most common reactive diluent, triethylene glycol dimethacrylate (TEGDMA), may also contribute to locally and systemically adverse effects from dental composites .
Upon polymerization, shrinkage stresses transferred to the tooth can cause deformation of the cusp or enamel microcracks , and stress at the tooth–composite interface may cause adhesive failure, initiation of microleakage and recurrent caries. In addition, significant increases in volumetric shrinkage and shrinkage stress are experienced when the double bond conversion is increased to reduce the leachable monomer . This trade-off between conversion and shrinkage has been an inherent problem with composite restorative materials since their inception. Thus, it is important to develop rapidly curing dental restorative materials with improved monomer conversion and mechanical properties that reduce volumetric shrinkage and shrinkage stress.
The challenge and potential benefits associated with the development of new resin systems can be demonstrated with the recent introduction of the silorane resin-based composite by 3 M/ESPE . The methacrylate-free resin matrix relies on cationic ring-opening polymerization of oxirane monomers, which may result in a reduction in polymerization stress primarily due to an inherently lower molar shrinkage value relative to methacrylates. The mechanical properties of the silorane composite are similar to those of dimethacrylate composites and acceptable for use as dental restorative materials.
Recently, thiol–enes have been investigated as alternatives to dimethacrylate-based RBCs . Methacrylate systems polymerize via a radical chain growth polymerization mechanism. In contrast, traditional binary thiol–ene systems utilize ene monomers that alone do not readily engage in chain growth polymerization. The thiol–ene reactions proceed via a radical step growth addition mechanism that comprises the addition of a thiyl radical to an ene functional group, followed by chain transfer to a thiol, thus regenerating the thiyl radical . Thiol–enes have advantageous curing kinetics, demonstrating rapid polymerization rates, high overall functional group conversion, and little sensitivity to oxygen inhibition . Most importantly for dental restorative materials, thiol–enes exhibit reduced shrinkage and a delayed gel point conversion . As a result of the delayed gel point, much of the shrinkage occurs prior to gelation, which dramatically reduces the shrinkage stress in the final polymer material by allowing flow to accommodate the pre-gel shrinkage. Unfortunately, these previous studies have also shown that traditional binary thiol–ene systems exhibit limited flexural modulus and strength relative to dimethacrylate resins.
Methacrylate–thiol–ene systems exhibit a reaction mechanism that is a combination of both step and chain growth polymerizations . Additionally, thiyl radicals exhibit an addition rate to methacrylates that is five times greater than the addition rate for allyl ethers . The result of the difference in thiyl radical addition rates is that the early stages of the reaction are dominated by methacrylate homopolymerization and chain transfer and the latter stages of the reaction are dominated by thiol–ene polymerization . The hybrid nature of the polymerization results in reduced shrinkage stress associated with chain transfer delaying gelation and the thiol–ene component remaining as a diluent throughout the initial stages of the reaction . Employing thiol–enes as reactive diluents also results in systems that combine many of the advantageous properties of both methacrylate and thiol–ene systems, while precluding the use of TEGDMA, which is rather hydrophilic and prone to leaching. The resulting methacrylate–thiol–ene resin systems exhibit equivalent mechanical properties for flexural modulus and strength, equivalent curing rates, increased methacrylate functional group conversion, and reduced shrinkage stress relative to the dimethacrylate control systems .
It is well known that for binary thiol–ene step growth polymerizations, the thiol and ene components must be present in a 1:1 stoichiometric ratio of functional groups to achieve complete conversion and maximize polymer mechanical properties . Thus, in all previous (meth)acrylate–thiol–ene studies, 1:1 thiol–ene stoichiometries were exclusively investigated . However, in the cases where both (meth)acrylate and ene functional group conversions were resolved in FTIR, lower ene conversions were reported for many of these systems . The lower ene conversion in ternary systems results from methacrylate and acrylate functional groups participating in reactions with both thiyl radicals and with methacrylate radicals . Therefore, in ternary systems once chain transfer to the thiol has occurred, the thiyl radicals are capable of reacting with either the ene or (meth)acrylate species, while the ene functional groups typically only react with thiol functionalities. Thus, when a 1:1 thiol:ene mole ratio is utilized in ternary systems, the thiol functional groups become a limiting reagent, resulting in relatively low overall ene functional group conversion. This behavior suggests that a 1:1 thiol–ene ratio would not be optimum for ternary (meth)acrylate–thiol–ene systems and those systems with thiol to ene stoichiometric ratios greater than one would increase the functional group conversions of each component. Additionally, by incorporating more thiol content into the reaction, additional chain transfer will be prevalent, resulting in further delayed gelation and reduced shrinkage stress.
The thiol–ene reaction has numerous inherent advantageous properties for dental restorative materials. One of the primary drawbacks of thiol–ene chemistry is the odor associated with thiol monomers. In the case of the systems being evaluated here, the tetrafunctional thiol monomer has a relatively high molecular weight (488.66 g/mol) and thus a mild odor relative to other lower molecular weight thiol materials. Additionally, when utilized as a reactive diluent, the overall composition contains relatively little thiol monomer further minimizing any potential concerns regarding odor. When further mixed with fillers (70 wt%), methacrylate–thiol–ene composites exhibit no distinguishable odor resultant from the thiol monomer content.
This study further evaluates thiol–enes, particularly in off-stoichiometric molar ratios, as reactive diluents in dimethacrylate resins. Two different ene monomers were utilized, a commercially available triallyl ether (TATATO) and a trinorbornene (TMPTN) that was synthesized as previously reported . Both of these monomers exhibited desirable mechanical properties and low shrinkage stress when evaluated previously in methacrylate–thiol–ene systems . Here, these systems were studied in greater detail by curing with visible light and examining the effects of varying thiol–ene content and stoichiometry. Polymerization kinetics and overall functional group conversion, flexural modulus and strength, and polymerization shrinkage stress were evaluated and compared to control dimethacrylate resins under identical curing conditions. It was hypothesized that increased thiol–ene content and increased stoichiometric ratio of thiol to ene functional groups will result in increased overall functional group conversions and decreased polymerization shrinkage stress, while maintaining flexural properties.
2
Experimental
2.1
Materials
The tri-ene monomer triallyl-1,3,5-triazine-2,4,6-(1H,3H,5H)-trione (TATATO) was purchased from Aldrich (Milwaukee, WI). The photoinitiator Irgacure 819 (I819) was donated by Ciba Specialty Chemicals (Tarrytown, NY). The inhibitor aluminum N-nitrosophenylhydroxylamine (Q1301) was donated by Wako Pure Chemicals (Osaka, Japan). 2,2-Bis[4-(2-hydroxy-3-methacryloyloxypropoxy)phenyl]propane (BisGMA), ethoxylated bisphenol-A dimethacrylate (EBPADMA), and triethylene glycol dimethacrylate (TEGDMA) were donated by Esstech Inc. (Essington, PA). Pentaerythritol tetra(3-mercaptopropionate) (PETMP) was donated by Evans Chemetics (Waterloo, NY). All chemicals were used as received. The norbornene monomer trimethylolpropane tri-(norborn-2-ene-5carboxylate) (TMPTN) was synthesized by the previously published procedures . Chemical structures of all monomers utilized in this study are shown in Fig. 1 .
2.2
Methods
Photopolymerization of samples for all analyses was conducted using 0.3 wt% I819 as the photoinitiator and 0.035 wt% Q1301 as inhibitor. Specimens were irradiated with 29 mW/cm 2 of light from an EXFO Acticure (Mississauga, Ontario, Canada) with a 400–500 nm filter. Irradiation intensity was measured at the sample surface level with an International Light, Inc. Model IL1400A radiometer (Newburyport, MA).
2.3
Fourier transform infrared spectroscopy (FTIR)
Kinetic analysis was conducted using a Nicolet 750 Magna FTIR spectrometer (Madison, WI) with a KBr beam splitter and an MCT/A detector. Series scans were recorded at a rate of approximately two scans per second until the reaction was complete, as indicated by the functional group absorption peak no longer decreasing. Experiments were conducted in the near infrared region (7000–4000 cm −1 ) with samples placed between glass slides with a 1.0 mm glass spacer. Functional group conversions were monitored utilizing the characteristic C–H absorption peaks at 6164 cm −1 (methacrylate) and 6132 cm −1 (allyl ether). These methacrylate and allyl ether peaks are overlapped in the near infrared region. Thus, a Gaussian fitting peak deconvolution method was utilized to determine individual functional group conversions. Norbornene functional groups did not exhibit a strong enough absorption in the near infrared region to accurately determine functional group conversion. For each composition, experiments were performed in triplicate.
2.4
Flexural modulus and strength
Samples measuring 2 mm × 2 mm × 25 mm were prepared using Teflon molds and cured under identical conditions as in the FTIR analysis. Polymer flexural strength and modulus were calculated using data obtained from a 3-point flexural test, carried out with a hydraulic universal test system (858 Mini Bionix, MTS Systems Corporation, Eden Prairie, MN, USA) using a span width of 20 mm and a crosshead speed of 1 mm/min. For each material, at least three replicates were evaluated.
2.5
Shrinkage stress
Experiments were performed with a tensometer (American Dental Association Health Foundation), which monitors stress development using cantilever beam deflection theory . Simultaneous conversion measurements were facilitated using remote near infrared transmitted through the specimen via fiber optic cables. Specimens were placed between 6 mm diameter glass rods with silane treated ends and measured 1.5 mm in thickness. The curing light was transmitted through the lower glass rod with a measured irradiation intensity of approximately 42 mW/cm 2 at the tip of the glass rod. For each composition, experiments were performed in triplicate.
2.6
Statistical analysis
The experimental results were analyzed using one-way analysis of variance (ANOVA) for FTIR conversion measurements, shrinkage stress, and flexural modulus testing. Multiple pair-wise comparisons were further conducted using Tukey’s test or the Tukey–Kramer protocol with a significance level of 0.05.
2
Experimental
2.1
Materials
The tri-ene monomer triallyl-1,3,5-triazine-2,4,6-(1H,3H,5H)-trione (TATATO) was purchased from Aldrich (Milwaukee, WI). The photoinitiator Irgacure 819 (I819) was donated by Ciba Specialty Chemicals (Tarrytown, NY). The inhibitor aluminum N-nitrosophenylhydroxylamine (Q1301) was donated by Wako Pure Chemicals (Osaka, Japan). 2,2-Bis[4-(2-hydroxy-3-methacryloyloxypropoxy)phenyl]propane (BisGMA), ethoxylated bisphenol-A dimethacrylate (EBPADMA), and triethylene glycol dimethacrylate (TEGDMA) were donated by Esstech Inc. (Essington, PA). Pentaerythritol tetra(3-mercaptopropionate) (PETMP) was donated by Evans Chemetics (Waterloo, NY). All chemicals were used as received. The norbornene monomer trimethylolpropane tri-(norborn-2-ene-5carboxylate) (TMPTN) was synthesized by the previously published procedures . Chemical structures of all monomers utilized in this study are shown in Fig. 1 .
2.2
Methods
Photopolymerization of samples for all analyses was conducted using 0.3 wt% I819 as the photoinitiator and 0.035 wt% Q1301 as inhibitor. Specimens were irradiated with 29 mW/cm 2 of light from an EXFO Acticure (Mississauga, Ontario, Canada) with a 400–500 nm filter. Irradiation intensity was measured at the sample surface level with an International Light, Inc. Model IL1400A radiometer (Newburyport, MA).
2.3
Fourier transform infrared spectroscopy (FTIR)
Kinetic analysis was conducted using a Nicolet 750 Magna FTIR spectrometer (Madison, WI) with a KBr beam splitter and an MCT/A detector. Series scans were recorded at a rate of approximately two scans per second until the reaction was complete, as indicated by the functional group absorption peak no longer decreasing. Experiments were conducted in the near infrared region (7000–4000 cm −1 ) with samples placed between glass slides with a 1.0 mm glass spacer. Functional group conversions were monitored utilizing the characteristic C–H absorption peaks at 6164 cm −1 (methacrylate) and 6132 cm −1 (allyl ether). These methacrylate and allyl ether peaks are overlapped in the near infrared region. Thus, a Gaussian fitting peak deconvolution method was utilized to determine individual functional group conversions. Norbornene functional groups did not exhibit a strong enough absorption in the near infrared region to accurately determine functional group conversion. For each composition, experiments were performed in triplicate.
2.4
Flexural modulus and strength
Samples measuring 2 mm × 2 mm × 25 mm were prepared using Teflon molds and cured under identical conditions as in the FTIR analysis. Polymer flexural strength and modulus were calculated using data obtained from a 3-point flexural test, carried out with a hydraulic universal test system (858 Mini Bionix, MTS Systems Corporation, Eden Prairie, MN, USA) using a span width of 20 mm and a crosshead speed of 1 mm/min. For each material, at least three replicates were evaluated.
2.5
Shrinkage stress
Experiments were performed with a tensometer (American Dental Association Health Foundation), which monitors stress development using cantilever beam deflection theory . Simultaneous conversion measurements were facilitated using remote near infrared transmitted through the specimen via fiber optic cables. Specimens were placed between 6 mm diameter glass rods with silane treated ends and measured 1.5 mm in thickness. The curing light was transmitted through the lower glass rod with a measured irradiation intensity of approximately 42 mW/cm 2 at the tip of the glass rod. For each composition, experiments were performed in triplicate.
2.6
Statistical analysis
The experimental results were analyzed using one-way analysis of variance (ANOVA) for FTIR conversion measurements, shrinkage stress, and flexural modulus testing. Multiple pair-wise comparisons were further conducted using Tukey’s test or the Tukey–Kramer protocol with a significance level of 0.05.
3
Results
3.1
Polymerization kinetics and resin conversion
Polymerization kinetics were monitored for the methacrylate–thiol–ene and dimethacrylate control resins under identical curing conditions. Results are shown in Fig. 2 and Table 1 . The methacrylate–thiol–ene systems all exhibited increased methacrylate conversion relative to the control dimethacrylate systems. For the off-stoichiometric systems containing the same overall weight percent of thiol–ene resin, the ratio of thiol to ene functional groups was 1.5:1, 2:1, or 3:1. As the ratio of thiol to ene functional groups was increased in the EBPADMA/PETMP:TATATO 70/30 system, the ene functional group conversion increased from 45 ± 1% for the 1:1 system to 69 ± 1% for the 3:1 system. The ene functional groups polymerize more slowly than the methacrylate functional groups. This behavior is due to the pseudo two-stage reaction between methacrylates and thiol–enes and is consistent with previously observed results for methacrylate–thiol–ene systems . Methacrylate functional group conversion also increased slightly from 92 ± 1% to 95 ± 1%. In the EBPADMA/PETMP:TATATO 60/40 system, the ene conversion reached 75 ± 1% for the 2:1 and 85 ± 1% for the 3:1 system and methacrylate functional group conversion increased to 97 ± 1% in the 2:1 and 98 ± 1% in the 3:1 system. The EBPADMA/PETMP:TMPTN also exhibited increased methacrylate functional group conversion as the thiol–ene ratio and content were increased. Norbornene functional group conversions were not resolvable. Kinetic results for norbornene containing resins in Figure 2 c demonstrate nearly equivalent polymerization rates for all of the systems that were evaluated as indicated by similar conversion-time lines for each of these functional groups.
