|Home | About | Journals | Submit | Contact Us | Français|
To investigate the long-term development of the post-cure degree of conversion (DC) for two flowable bulk-fill composites.
Tetric EvoFlow Bulk Fill (TEFBF) and SDR were chosen due to their distinct compositional modifications that enable the decrease of translucency during polymerization and lower polymerization rate, respectively. DC was assessed using FT-Raman spectroscopy at the post-cure times of 0 h, 24 h, 7 d and 30 d. The post-cure behavior was analyzed by a mixed model ANOVA and partial eta-squared statistics.
DC ranged from 61.3-81.1% for TEFBF and 58.9-81.6% for SDR. The initial (0 h) DC was significantly lower at a depth of 4 mm than at a depth of 1 mm (4.9% for SDR and 11.1% for TEFBF). Both materials presented a significant post-cure DC increase, up to 16.4% for TEFBF and 20.6% for SDR. The post-cure DC development was depth-dependent for TEFBF, but not for SDR. The post-cure DC increase was observed during 24 h for TEFBF and 7 d for SDR.
Some of the bulk-fill composites may need longer times than the commonly accepted 24 h to reach the final conversion. This may be attributed to their compositional modifications that are mostly undisclosed by manufacturers. Our findings imply that investigations commonly performed 24 h post-cure may underestimate some of the bulk-fill composite properties, if these are affected by the slowly-developing DC. Reactive species may also be available for leaching out of the restoration during an extended time period, with possible implications on biocompatibility.
The polymerization reaction of dental composites features complex kinetics and is limited by mobility restrictions imposed on reactive species during the growth and crosslinking of polymeric chains (1). Although most of the polymerization occurs during the light curing, commonly lasting for about 20 s in clinical practice, the post-cure polymerization slowly furthers after the light curing has ended (2). This phenomenon is caused by the reduction of polymerization rate due to an immense increase in viscosity of the reaction medium. Very low mobility causes the reaction rate to drop for several orders of magnitude (3) while the considerable amounts of reactive species (free radicals and methacrylate monomers) are still available (4). Thus the post-cure polymerization continues at very low and decreasing rate, until the viscosity increase completely immobilizes the remaining reactive species. This causes the polymerization to stop before the reactants are completely consumed, for dental composites typically at the DC values of 60-80% (1, 5, 6). The post-cure polymerization is generally considered to be completed within the 24 h after light curing (7-9). Although the literature does not explicitly claim that the post-cure polymerization ceases within 24 h, this is implicitly suggested by the design of numerous in vitro studies that are performed after an aging period of 24 h, which is considered sufficient to account for the changes in polymeric network structure due to the post-cure polymerization (7, 8, 10-13). The post-cure polymerization beyond 24 h has been scarcely investigated; only one study reported the DC increase after 30 days post-cure for composite samples that received lower radiant energies (14). Another study suggested the post-cure polymerization beyond 24 h by monitoring the gradual increase in hardness for up to 7 days (15).
A new generation of dental composites termed “bulk-fill” was recently introduced to the market and has been increasingly accepted due to a shortened and simplified clinical procedure. The bulk-fill composites are designed for placement in layers of 4-5 mm, opposite to conventional composites that enable maximum layer thickness of 2 mm. This is made possible by addressing two main factors that limited the composite layer thickness: polymerization shrinkage stress and low translucency (16, 17). The shrinkage stress is mitigated by lowering the elastic modulus (13), whereas the translucency is enhanced by lower filler ratio and larger particle size (17). Apart from these modifications, the fundamental chemistry of bulk-fill composites does not considerably differ from that of conventional composites. Their basis remains the photocurable blend of bifunctional methacrylate monomers and silanized glass fillers. However, there are some bulk-fill composites which exhibit innovative approaches for improving clinical performance. Namely, Tetric EvoFlow Bulk Fill (TEFBF) changes its translucency from 28% to 10% during polymerization in order to improve the esthetic appearance by mimicking the dentinal opacity (18). Another example is SDR that contains a proprietary modified urethane-dimethacrylate resin containing photoactive groups in order to control polymerization kinetics and retard the buildup of polymerization shrinkage stress (19). The aforementioned approaches may affect both the short term (during light curing) and long term (post-cure) polymerization behavior (19). In other words, the novel technologies employed in TEFBF and SDR may influence the DC immediately after curing, as well as the amount and duration of the post-cure DC increase. This means that the post-cure polymerization might last longer than 24 h, as commonly believed. Additionally, light curing of thick increments might cause lower initial DC of the bottom layer (12) and possibly a post-cure polymerization of higher extent or longer duration.
The aim of this study was to investigate the post-cure DC development of two composites TEFBF and SDR at depths of 1 mm and 4 mm at post-cure times of 0 h, 24 h, 7 d and 30 d. The null hypotheses assumed no difference in:
Detailed data on the investigated bulk-fill composites are shown in Table 1. Four cylindrical samples (d = 3 mm, h = 5 mm) were made for each composite as well as the measuring depth using a custom-made stainless steel split-mold. Uncured composite paste was placed into the mold, both mold apertures were covered with a polyethylene terephthalate (PET) film and curing was performed for 20 seconds through the upper aperture with the LED curing unit (Bluephase G2, Ivoclar-Vivadent, Schaan, Liechtenstein); wavelength range 380-515 nm, irradiance 1185 mW/cm2, as measured with integrating sphere (IS, Gigahertz Optik GmbH, Puchheim, Germany). The curing unit tip was positioned directly on top of the PET film covering the sample and the environmental temperature during curing was 21±1 °C. Raman spectra were collected from the depths of 1 mm and 4 mm at four time points: 0 h (immediately after light-curing), 24 h, 7 d and 30 d, representing repeated measurements on the same samples. The samples were dark stored in the incubator (Cultura, Ivoclar-Vivadent, Schaan, Liechtenstein) at 37±1 °C between the individual measurements at the mentioned time points.
A FT-Raman spectrometer Spectrum GX (PerkinElmer, Waltham, USA) was used for DC evaluation, as previously described (5). Briefly, Raman scattering was induced by an NdYaG laser (1064 nm) with power of 400 mW. The spectra were collected from the surface of cylindrical samples at selected depths by exposing the spot of 0.5 mm in diameter to the excitation beam. The spectral resolution was 4 cm-1 and 50 scans were taken for each spectrum. The spectra of the uncured composites (n=4) were collected using the same parameters. DC was calculated as the relative intensity change of the aliphatic C=C band at 1640 cm-1 normalized to the aromatic C=C band at 1610 cm-1, according to the equation: DC = 1–Rpolymerized/Runpolymerized, where R = (aliphatic C=C peak height) /(aromatic C=C peak height).
Since the data were normally distributed, they were summarized as mean values ± standard deviations. Mean DC values were compared by a mixed model ANOVA, with “material” and “depth” as the between-subject factors and “time” as the within-subject factor. Partial eta-squared statistics was used to assess the relative influences of the factors and their interactions. Multiple comparisons were performed using Tukey’s HSD and Bonferroni adjustments for independent and dependent observations, respectively. Statistical analysis was performed in SPSS 20 (IBM, Armonk, NY, USA) with α=0.05.
Mean DC values and results of statistical analysis are shown in Figure 1, whereas measures of effect size (partial eta-squared values) are shown in Table 2. The DC values ranged from 61.3-81.1% for TEFBF and 58.9-81.6% for SDR. The initial (0 h) DC was significantly lower at the depth of 4 mm than at the depth of 1 mm for both materials (4.9% for SDR and 11.1% for TEFBF).
A statistically significant interaction was observed for the factors time*material and time*depth, indicating differences in post-cure DC increase between the materials and measuring depths. Hence, the influence of the factors “time” and “depth” was analyzed for each material separately (Table 2). Both composites showed a significant influence of the factors “time” and “depth”, with higher influence of the factor “time”. Additionally, TEFBF showed a significant interaction of the factors “time” and “depth”, indicating that the post-cure development of DC differed between the measuring depths of 1 mm and 4 mm.
Both materials presented a significant post-cure DC increase, amounting up to 16.4% for TEFBF and 20.6% for SDR. At the depth of 4 mm, SDR also showed a statistical heterogeneity of DC values at 0 h, 24 h and 7 d, indicating that the post-cure DC increase could be detected for up to 7 days post-cure. Also, for SDR at 1 mm depth, a non-significant trend of DC increase for up to 7 days was observed.
Both TEFBF and SDR at 30 days post-cure presented high DC values ranging between 76.1-81.6%, despite the significant initial differences. The DC values reached at time points of 7 d and 30 d were statistically similar for both materials at both measuring depths.
This study investigated a long term DC development of two bulk-fill composites at two depths. The first null hypothesis was rejected, since the DC values increased with time for both materials and depths. By comparing the post-cure DC increase within a single material and depth, it can be seen that TEFBF reached the DC plateau after 24 h at both measuring depths, while SDR showed a significant DC increase for up to 7 days post-cure at 4 mm depth (Figure 1). This suggests that deeper layers of SDR need longer times than the commonly accepted 24 h (7, 8, 12) to attain their final DC. The prolonged post-cure polymerization found in SDR may be attributed to its modified urethane dimethacrylate that incorporates photoactive groups as a strategy to decrease shrinkage stress development through controlling of the polymerization kinetics (19). In the manufacturer brochures, this is referred to as a “polymerization modulator” and is claimed to favor linear chain growth instead of crosslinking (20). Such behavior allows for more viscous flow and acts to decrease the shrinkage stress during polymerization of SDR (19). The side-effect of the polymerization modulator might have been the retardation of conversion kinetics, resulting in extended post cure polymerization. Also, the presumably lower crosslinking density of the polymeric network might allow better mobility of reactive species (21) thus supporting the prolonged post-cure polymerization. The trend of post-cure DC increase for up to 7 days was also observed for SDR at 1 mm depth, but there was no statistical significance. The inability of our analysis to detect statistical significance could be attributed to the lower extent of post-cure polymerization occurring at 1 mm depth, due to higher initial DC. Since the curing light was much less attenuated at the depth of 1 mm than at 4 mm (22), more monomers were consumed during light-curing and thus lower amount of monomers was left for the subsequent post-cure polymerization (23).
Statistical analysis showed a significant interaction between the factors “time” and “depth” for TEFBF (Table 2), indicating that the development of DC through time was different between depths of 1 mm and 4 mm, thus leading to rejection of the second null-hypothesis in the case of TEFBF. This difference is mostly caused by lower initial DC and higher extent of post-cure polymerization at the depth of 4 mm (Figure 1a). The light passing through the material is attenuated due to absorption and scattering at filler particles (22), hence the depth of 4 mm received lower radiant energy. According to the Lambert-Beer law, the light intensity decreases exponentially with length of the light path (24). As a result, the number of activated photoinitiator molecules is diminished as the layer thickness increases, reflecting in significantly lower initial (0 h) DC (61.3%) at 4 mm, compared to that at 1 mm (72.4%). However, this difference was leveled after 24 h (Figure 1a), suggesting that the radiant energy delivered to 4 mm enabled sufficient extent of post-cure polymerization to compensate for initially lower DC. This is possible because the methacrylate polymerization occurs via free-radical addition reaction that preserves a free radical at the end of polymeric chain throughout the propagation (25). In this way, a single free radical can trigger addition of a large number of monomer units before it becomes inactivated by termination or mobility restrictions. In fact, the extent of polymerization in filled dimethacrylate resins is mainly determined by mobility restrictions caused by an enormous increase in viscosity rather than the depletion of reactants (26). Therefore, the number of free radicals is not the main limiting factor for extent of polymerization, as evidenced by their persistence after the end of polymerization (4). This may explain our results for TEFBF, which showed that the amount of free radicals at 4 mm depth was sufficient to attain statistically similar post-cure DC to that at 1 mm depth, despite the significant initial differences.
In contrast to TEFBF, SDR showed similar post-cure DC development at depths of 1 mm and 4 mm, as indicated by no significant interaction of factors time*depth (Table 2). Therefore, the second null-hypothesis was accepted for SDR. It appears that the high intensity of curing light (1185 mW/cm2) and curing time of 20 seconds combined with high translucency of SDR (17) activated enough photoinitiator molecules to result in statistically similar post-cure conversion rate at both depths. Additionally, the translucency of SDR increases during polymerization due to the decrease of refractive index mismatch between the resin and filler particles and a consequent decline of light scattering (17). This facilitates curing efficiency at higher depths, since the light transmission through the overlying layers improves as they are cured.
Although the initial (0 h) DC values were significantly lower at 4 mm than at 1 mm for both composites, the difference for SDR (4.9%) was twice lower than that of TEFBF (11.1%). This may have been caused by high translucency of SDR (17), as well as a unique translucency change of TEFBF (18). TEFBF contains a patented technology of finely tuned refractive indices of resin and filler particles, allowing for high translucency of the uncured paste (28%) that progressively decreases during polymerization to about 10% (18). The transition from high to low translucency allows high light penetration during the initial phase of light curing and an esthetic appearance due to low translucency of a cured restoration. In this regard, TEFBF behaves opposite to most of the commercial composites (including SDR), whose translucency increases during curing (27, 28). Since the upper parts of the TEFBF samples received higher irradiance than deeper layers, they underwent the translucency decrease and consequently attenuated the light delivered to the 4 mm depth. This could explain the relatively large DC difference between the depths of 1 mm and 4 mm in the case of TEFBF. However, despite the 4 mm depth exibited an initially lower DC, the post-cure polymerization completely negated the difference, yielding statistically similar DC values between 1 mm and 4 mm at all post-cure times (24 h, 7 d and 30 d).
Considering the initial (0 h) DC values at 1 mm depth, TEFBF showed significantly higher values than SDR. The initial DC is determined by a complex interplay of multiple parameters such as structure, reactivity and relative amounts of monomers (25), filler load and geometry (29, 30), photoinitiator chemistry (31) and proprietary undisclosed additives (18, 19). In consequence, the direct comparison of DC values between TEFBF and SDR does not provide much indication of the overall material performance. The DC data alone are insufficient to compare the DC-dependent properties among different formulations, since these properties are considerably influenced by a number of other factors (32). However, it is interesting to note that these two composites, each with its own proprietary modifications and different initial DC values, managed to converge to statistically similar DC values at both depths after 7 days post-cure (Figure 1). Therefore, the third null hypothesis was accepted for post-cure times of 7 d and 30 d. This suggests that both composites have similar “maximum attainable” DC, that is reached after a certain post-cure time, irrespective of the initial DC values and differences in polymerization rate. This final DC was probably determined by the mobility of unreacted species, which was influenced by filler load (29), geometry and surface treatment (33), as well as resin viscosity (34). This cannot be discussed into detail, since the compositional information is only partially disclosed (Table 1). As both composites contain filler load of about 45 vol%, it could be speculated that it was the major determinant of final DC values (29, 30), while other factors were probably less important. Also, an alternative explanation is that other DC-determining factors might have been similar in both composites.
Throughout the polymerization of multifunctional methacrylates, the cyclization competes with crosslinking (35) and the latter becomes dominant as the polymerization advances. Thus, even a small DC increase at late phases of polymerization might considerably enhance the crosslinking density of a polymeric network (26, 36). Despite the DC improvement that occurred in SDR beyond 24 hours was small relative to the DC attained within the first 24 h post-cure (about 6% and 75%, respectively), it may favorably reflect on other material properties, mainly those dependent upon crosslinking, i.e. mechanical properties (26), water sorption, degradation in aqueous environment and release of leachable components (37).
This study demonstrated that some of the chemistries used in bulk-fill composites might cause prolonged post-cure DC increase of up to 7 days. The materials TEFBF and SDR were chosen because they feature two novel technologies that are not contained in other commercial composites. Our findings must not be generalized. They should not be applied to the whole class of bulk-fill composites because each material features different composition and unique modifications to achieve the bulk-fill capability. However, these modifications may inadvertently affect other important material properties; hence their effects should be further investigated.
The finding of prolonged post-cure polymerization has two main implications. First, the post-cure times at which the DC-dependent composite properties are tested in laboratory setting may need to be longer than the commonly accepted 24 hours since the properties of some bulk-fill composites may develop beyond this time. Second, the longer post-cure polymerization suggests that reactive species are mobile and available for leaching into the oral environment during an extended time period, which raises biocompatibility concerns.
This investigation was supported by Croatian Science Foundation (Project 08/31 Evaluation of new bioactive materials and procedures in restorative dental medicine). The Kinetics add-on for Matlab was kindly provided by Professor E. Goormaghtigh.
The authors declare no conflict of interest.