Effect of 2,3-Butanedione in an Experimental Dentin Bonding System on Degree of Conversion and Microleakage of Class V Composite Restorations

Recent advances in dental adhesives and composite resins are responsible for the popularity of these materials for direct restoration of carious lesions [1]. At present, bond to dentin is weaker than the bond to enamel due to the heterogeneous and moist nature of dentin. Thus, a long-lasting, durable bond to dentin is less commonly achieved compared to enamel [2,3]. Of all the components involved in bonding at the resin-dentin interface, the adhesive layer has the lowest modulus of elasticity [4,5]. Thus, most adhesive systems available in the market contain fillers that increase the mechanical properties and the bond strength of the restorations [6-9]. Received: August 12, 2019

Nanometer-sized fillers like clay are available; By performing adequate surface treatments, these fillers can be lodged in between polymer chains in dentin adhesive layer and increase the mechanical properties of the adhesive and subsequently enhance the bond strength [10,11].
Addition of clay nanoparticles to dentin bonding agents results in fast deposition of particles due to their higher density. However, this problem can be solved by surface treatment of nano clay particles with a polymer, which decreases density and results in a reaction between active groups bonded to the surface of particles and the solvent. One effective method to form active groups on the surface of nanoparticle is their grafting with monomers [4,5,12].
Studies on this topic have demonstrated that grafting nano-clay particles with poly acrylic, poly methacrylic and poly methyl methacrylic acids enhances the mechanical properties of adhesives [12][13][14].
Using clay nanoparticles grafted with polyacrylic acid in the recent studies has shown significant improvement in bond strength [14]. Incorporation of these materials into the experimental dentin bonding agent suggests the theory that adhesives containing active acidic groups deactivate the amine group and decrease the degree of conversion of adhesives with CQ/amine photo initiator resulting in a weak bond between dentin and composite and allowing higher diffusion of water at the composite/dentin interface and subsequently higher microleakage. To overcome this problem, a single step photo initiator like 2,3 butanedione can be used.
In the recent studies, several types of new photo initiators such as 2,3 butanedione, 1-phenyl-1,2-propanedione and hexafluoroantimonate p-octyloxy-phenyl iodonium have been evaluated [15,16]. CQ is the most commonly used photo initiator in light-cure composite resins [15]. CQ is naturally yellow causing problems in color matching. This limits the application of its higher concentrations and consequently, the DC and curing depth [15]. The 2,3 butanedione (Diketone BD) is a viscous liquid. Due to its physical state, it is well compatible with resin components and in contrast to CQ, which is solid at room temperature, it can be used as a diluting agent in composite resins. The 2,3 butanedione is a photo initiator without the amine activator. By light irradiation and cleavage of C-C bond, it produces two carbonyl free radicals and initiates polymerization [15,17]. In the UV -VIS spectrum, butanedione has a peak absorbance at 419 nm, which is close to the output power of dental light curing units. On the other hand, shorter absorbance wavelength of butanedione lowers its yellowness compared to that of CQ [15]. The aim of the present study was to evaluate the effect of type of photo-initiator (2,3 butanedione, CQ/amine or both) in an experimental dentin bonding agent on the DC of adhesive and microleakage of class V composite restorations bonded with this experimental dentin bonding agent.

Materials and Methods
The graft polymerization of acrylic acid onto the surface of nanoclay was carried out according to the previous works [14].
The grafted nanoclay powder was added to the adhesive solution and the obtained mixture was sonicated for 3 minutes under 50% power and 6 pulses (Each pulse for 1 second). Dentin adhesive system was manufactured in the Iranian Polymer Research Center according to the formulation presented in Table 1. Polyacrylic acidgrafted nano clay filler particles were added by 0.2wt%. Photo initiator was added as follows:

Measurement of Microleakage
Specimens were subjected to 3000 thermal cycles using digital thermocycler (Willytec version 3  Significance level was set at 0.05.

Results
Data for scoring of microleakage in the study groups is reported in (p=0.046), and between groups 2 and 3 (p=0.046). However, the difference between groups 1 and 3 was not significant (p=0.055).  bKruskal-Wallis test. Table 3: Score of microleakage at the occlusal/gingival margin of restorations in the study.  Note: * Kruskal -Wallis

Discussion
Microleakage is a factor responsible for pulpal reactions to restorative materials. Formation of microscopic gaps at the toothrestoration interface enables the passage of ions, molecules, bacteria and their toxic products as well as liquids, causing postoperative tooth hypersensitivity, marginal discoloration, secondary caries and pulp inflammation [19][20][21]. Degree of conversion (DC) is an important criterion for clinical success of a composite resin restoration. Clinically, polymerization of materials used in esthetic dentistry must be increased as much as possible to achieve the maximum restoration strength. Although increased DC increases the polymerization shrinkage of adhesive or restoration and subsequently imparts tensile stresses to the bonded interface, it can be compensated by application of suitable clinical technique; while inadequate polymerization causes problems such as less biocompatibility, lower physical and mechanical properties and early failure of the restoration [22].
In the current study, a filler-containing dentin adhesive was used. Studies have reported that filler-containing adhesives have improved mechanical properties and enhance the bond strength [6][7][8][9]13,14]. Kim et al. [23] reported that addition of less than 1wt% nanofillers to dental adhesives increased the microtensile bond strength of the composite restoration; but in higher amounts of nanofillers, the microtensile bond strength decreased due to the aggregation of particles and formation of clusters. Solhi et al. [14] demonstrated that dental adhesives containing 0.2 wt% PAA-g-nano clay particles, showed a significantly increased bond strength.
The photo initiator is often added in the range of 0.1-1% [22]. In the present study, we applied 1-1.5wt% of photo initiator within the formulation of the experimental adhesive.
There are some relevant methods to determine the degree of monomer to polymer conversion of composite resin specimens like FTIR spectroscopy, Laser Raman spectroscopy, Electron Spin Resonance (ESR), Infrared Spectroscopy (IR), Dynamic Mechanical Thermal Analysis (DMTA), Attenuated Total Reflection (ATR) etc [24]. In the current study, FTIR was used because it is a beneficial method for molecular detection and tracing of chemical reactions.
It provides a better interpretation of the changes that occur in the polymer matrix. FTIR can easily detect and quantify chemical bonds [25]. This method has been used by Solhi et al. [14], Park et al. [26], Yazdi et al. [27] and Shin et al. [16] to assess the polymerization efficacy of Bis-GMA-based composite resins. Dye penetration is the most commonly used technique for assessment of microleakage. This method has acceptable techniqual sensitivity and is easier than other methods available for this purpose [28].
Thus, researchers use fuchsin or methylene blue with particles equal in size or smaller than the size of bacteria (about 2 microns) [19]; In the current study, 10% methylene blue was used. The only drawback of this method is its grading system because its sensitivity decreases by re-evaluation after a time interval [19].
In-vitro studies on microleakage are often performed on class V restorations. Specimens are usually subjected to thermocycling or mechanical loading to simulate the oral environment [28]; In the current study, specimens were subjected to 3000 thermal cycles prior to microleakage assessment. The results showed that in all groups some degrees of microleakage existed and none of the formulations evaluated in this study could provide complete marginal seal. No significant difference was seen among the study groups in terms of microleakage.  Fortin et al. [29] compared microleakage in several dental adhesive systems and reported that some degrees of microleakage were detected in all groups; which is in accord with the current study results. According to a study by Mousavinasab et al. [18] score of microleakage was not significantly different in several experimental filler-containing adhesives and Single Bond. This finding is in line with our results. Our study also showed that in all groups, microleakage at the gingival margin (Figures 2-5) was significantly higher than that in the occlusal margin; This indicates that marginal seal in the dentin margin was lower than that in the enamel margins. Lower microleakage in the enamel compared to dentin margin is due to the stronger bond to enamel because the enamel has a higher mineral content and is more uniformly mineralized compared to dentin. Also, dentinal fluid does not exist in enamel and thus, monomers can more easily penetrate its microporosities created following etching, resulting in a more reliable micromechanical bond. Another reason for the lower microleakage in the enamel margin is beveling the enamel margin increases the bondable enamel surface [20].  Thus, in combination of CQ and 2, 3 butanedione, the reaction for generation of free radicals with butanedione is usually done with CQ even when amine is not present in the formulation.
In the current study, incorporation of a combination of butanedione and CQ showed more favorable results. Thus, future studies are suggested to assess the properties of higher concentrations of butanedione in combination with CQ. Moreover, since the effect of duration of solvent vaporization was not investigated in the current study, further studies are required to assess the effect of this factor.

Conclusion
Under limitations of the present study, Substitution of 1wt% butanedione for 1wt% CQ/amine in the studied experimental adhesive system or the combination of butanedione and CQ had no effect on the microleakage of class V composite restorations. Also, DC of adhesive containing 1% butanedione was significantly lower than that of the adhesive containing 1% CQ or combination of both.