Specimen Preparation and Conditioning

The materials and specimen preparation had been described in our earlier work.¹⁰ RBCs evaluated were comprised of three conventional RBCs (Filtek Z350 [FZ], Tetric N Ceram [TN], and Beautifil II [BT]), three bulk-fill restorative RBCs (Filtek Bulk-Fill Restorative [FB], Tetric N Ceram Bulk-Fill [TB], and Beautifil Bulk-Fill Restorative [BB]) along with three bulk-fill flowable RBCs (Filtek Bulk-Fill Flowable [FF], Tetric N Flow Bulk-Fill [TF], and Beautifil Bulk-Fill Flowable [BF]). The materials and their technical profiles are specified in Table 1. Briefly, 40 beam-shaped specimens (12×2×2 mm) of the different RBCs were fabricated with customized stainless-steel molds. The RBCs were light polymerized using an LED curing light (Demi Plus, Kerr Corp, Orange, CA, USA) with an output irradiance of 1330 mW/cm² in two overlapping irradiations of 10 seconds. Specimens were then finished using fine polishing discs (Sof-Lex, 3M ESPE, St Paul, MN, USA), inspected for defects/parallelism, and measured with a digital caliper (Mitutoyo Corp, Kawasaki, Japan). The RBCs were then randomly divided into four groups (n=10) and stored in sealed containers with the following conditioning mediums for seven days at 37°C in an incubator (IN-460, Memmert, Schwabach, Germany): air, artificial saliva (SAGF),¹⁷ 0.02 N citric acid, and 50% ethanol-water solution. Air served as the control medium, while artificial saliva replicated natural saliva. Citric acid and 50% ethanol mimicked certain fruits, candies, and beverages, including alcohol. The pH of the artificial saliva (6.8) and citric acid (2.6) was verified via a digital pH meter (pH2700, Eutech, Singapore).

Table 1 Technical Profiles and Manufacturers of the Materials Evaluated

View Fullscreen

Dynamic Mechanical Analysis

Following conditioning, the RBCs were subjected to dynamic flexural testing (DMA RSAG2, TA Instruments, New Castle, DE, USA) in the various conditioning mediums at 37°C. Details of the testing protocol have been described previously.¹⁶ Specimens were loaded using a three-point bending fixture with a 10-mm support distance and loading frequency of 0.1 to 10 Hz to simulate the range from almost static to the upper limit of chewing rates.¹⁸ However, the upper limit is extremely fast and exceeds the range expected for normal chewing. Loading frequency was progressively increased from 0.1 to 10 Hz in 15 steps with a delay time of one second and a delay cycle of 0.5 Hz. Storage modulus, loss modulus, and loss tangent data were subsequently calculated as follows:^16,19

Storage modulus:

Loss modulus:

Loss tangent:

where σ° is the maximum stress at the peak of the sine wave, ɛ° is the strain at the maximum stress, f_o is the force applied at the peak of the sine wave, b is the sample geometry term, and k is the sample displacement at the peak.

Sample geometry b for a three-point bending bar was calculated as follows:

where B is the width of the specimen (in mm), H is the height of the specimen (in mm), and L is the distance between the supports (in mm).

Statistical Analysis

Data were analyzed using the SPSS statistical program (version 12.0.1, SPSS Inc, Chicago, IL, USA). Parametric analysis was performed, as data were normally distributed based on the Shapiro-Wilk test. Two-way analysis of variance (ANOVA) was used to determine interactions between independent variables (materials and conditioning mediums) and dependent variables (storage modulus, loss modulus, and loss tangent). One way ANOVA followed by the post hoc Tukey test were used to assess intermedium and intermaterial differences. The latter was grouped by both manufacturer and material type (ie, conventional, bulk-fill restorative, and bulk-fill flowable). Correlation between viscoelastic properties was carried out with the Pearson correlation. All statistical analyses were performed at a significance level of α = 0.05.

Comparison Between Conditioning Mediums

In the present study, exposure to aqueous solutions degraded the viscoelastic properties of the RBCs with the exception of FB and FF. Findings corroborated previous studies pertaining to the effect of conditioning mediums/food-simulating liquids on viscoelastic properties of RBCs.^21,22 The lowest storage modulus and highest loss tangent were observed after conditioning in ethanol. This could be attributed to the high propensity of the RBC polymers to absorb ethanol-water solutions, resulting in matrix swelling and decreased mechanical properties.^23,24 The better performance of FB and FF may be attributed partly to the introduction of methacrylates that undergo free-radical addition fragmentation. This might enhance postcuring structural stability in addition to reducing polymerization stress.²⁵ For manufacturer C, no significant difference in storage modulus was observed between conditioning in citric acid and ethanol for all material types. All the products of manufacturer C were giomers based on prereacted glass ionomer (PRG) technology, where fluoride-containing glass is reacted with polyacid acids, freeze-dried, milled, silanized, and used as fillers.² Giomers have been found to be particularly prone to degradation by acids of low pH due to their glass ionomer constituent.²⁶

Comparison Between Material Types

When material types were compared, bulk-fill restorative and flowable RBCs generally had a significantly lower storage modulus than their conventional counterparts with the exception of FB. FB was significantly more rigid than both conventional (FZ) and bulk-fill flowable (FF) materials from manufacturer A. As the filler content by volume of FB was lower than FZ, improvements in mechanical properties can be attributed more to resin modifications. The bulk-fill restorative and flowable RBCs of manufacturers B and C were significantly less rigid than their conventional equivalents. Accordingly, their conventional RBCs should be used preferentially over bulk-fill materials, especially in high-stress–bearing areas. The storage modulus of the bulk-fill restoratives TB and BB were similar or even lower than their flowable equivalents TF and BF. This was unexpected, as the filler content of flowable RBCs is lower than restorative ones to reduce viscosity and enhance cavity adaptation. Previous studies based on static techniques found that bulk-fill RBCs have mechanical properties between conventional and flowable hybrid materials.^6-8 It is plausible that bulk-fill flowable RBCs are formulated similarly to flowable hybrid materials, resulting in comparable mechanical properties. The depth of cure and degree of conversion of bulk-fill flowable RBCs are also typically greater than bulk-fill restorative materials due to their greater translucency.^27-29 The latter is achieved via decreasing light absorption by pigments and light refraction by matrix-filler interfaces (arising from lower filler content). Degree of conversion governs mechanical properties (with higher degree of conversion being associated with better mechanical properties) and explains in part the similar rigidity of bulk-fill restorative and flowable RBCs in the present study.³⁰

Although no obvious trends were observed for manufacturer A, the two conventional RBCs for manufacturers B and C usually had significantly higher loss modulus and flowed more than their bulk-fill counterparts. As filler content of the conventional RBCs was higher or similar to the bulk-fill restoratives, observations can be accounted for only by variance in initiators and resin constituents. While TN is based on camphorquinone, both TF and TB incorporate an adjunct photoinitiator, Ivocerin, giving them a wider absorption range of 370 to 460 nm.³¹ As initiator systems influence both the degree of conversion and cross-linking, it is not surprising that flow is reduced in both TF and TB.³² For BF and BB, the supplementary copolymers added included urethane dimethacrylate (UDMA) and bisphenol-A polyethoxy-dimethacrylate. The lower flow of both giomers can be credited to UDMA, which is a rigid and viscous material, and possibly the use of PRG fillers.³³ As loss tangent values are dependent on both storage and loss modulus, trends are expected to vary, depending on viscoelastic changes after conditioning in the various mediums.

Comparison Between Products

When conventional materials were compared, BT was found to be significantly stiffer than the other two conventional materials regardless of conditioning medium. This may be due primarily to the higher volume filler loading of BT, including PRG fillers. Likewise, the greater loss modulus and tangent of TN can be accounted for by its relatively lower filler content. Among the bulk-fill restoratives, TB was significantly less rigid than FB and BB. As filler content was similar, the disparity could be due to differences in resin composition and the eclectic mix of fillers used in TB. The lower storage modulus of TB could also be contributed by the nonuse of a polywave curing light that is necessary for the photoactivation of Ivocerin. No obvious patterns were observed for loss modulus and tangent with ranking varying for different conditioning mediums. This reiterates the need for RBCs to be evaluated in different chemical environments instead of just air and/or distilled water, which is often practiced. When comparing the bulk-fill flowable materials, BF, in view of its high filler volume content, was significantly more rigid than TF and FF. As loss tangent is inversely proportion to rigidity, the significantly lower loss tangent of BF was anticipated in view of its higher storage modulus.

Correlations Between Viscoelastic Properties

As RBCs exhibit both viscous and elastic characteristics under stress, some degree of correlation is anticipated between storage and loss modulus. The correlation was moderately strong, supporting the need for routine viscoelastic characterization of RBCs in conjunction with static testing that evaluates only the elastic component of materials.^10-12 Significant correlations between storage/loss modulus and loss tangent were anticipated, as loss tangent is the ratio of loss modulus to storage modulus. As the correlation between storage modulus and loss tangent was substantially stronger than that between loss modulus and loss tangent, any variance in energy-dissipating capacity is due primarily to changes in rigidity. This corroborated previous work that reported loss tangent increases with increasing filler content.³⁴

Limitations of the Present Study

Some of the limitations of this study had been discussed in our earlier work.¹⁰ They included the need for longer-term conditioning in the various mediums and independent appraisal of rheological properties and filler content instead of reliance on manufacturers' data, which may not be comparable due to methodological differences. In addition, the evaluation of products from more manufacturers is warranted given the heterogeneity of products under the same material type or classification. With regard to DMA, future work could encompass variation of frequencies, temperatures, and displacement that simulate the range of forces and temperatures encountered intraorally. The glass transition temperature (T_g), which is the temperature at which RBCs undergo transition from a glassy to a rubbery state, can also be determined with DMA. This represents the point at which marked changes in physical properties occur and had been linked to curing light energy density and degree of monomer-to-polymer conversion.³⁵ T_g values may provide valuable insights toward explaining the observed viscoelastic responses. Viscoelastic properties should also be interrelated with the clinical performance of bulk-fill and conventional RBCs.