Physicochemical, Mechanical, and Esthetic Properties of the Composite Resin Manipulated with Glove Powder and Adhesive as a Modeling Liquid

Composite resins with low flowability are usually handled and manipulated before insertion into the tooth preparation with gloved hands and/or using an instrument covered with a little amount of adhesive to facilitate modeling. We investigated if the modeling techniques (combined or not) affected physicochemical and esthetic properties of a composite resin. Specimens were fabricated and divided into groups according to the handling/modeling technique: Gloved-hands (composite was hand-manipulated with powdered latex gloves); Adhesive (adhesive was used in between the composite layers); Gloved-hands + Adhesive; Control (no adhesive and no touch with gloved-hands). The highest values for flexural strength (MPa), modulus of elasticity (GPa), and fracture toughness (MPa.m0.5) were obtained for Adhesive and Gloved-hands + Adhesive (p < 0.05); the lowest values were obtained for Control and Gloved-hands (p < 0.05). The Control group had the highest sorption. The Gloved-hands (p < 0.05) group had the highest solubility. Adhesive and Gloved-hands + Adhesive had a similar solubility (p > 0.05). The Control group (p < 0.05) had the lowest solubility. There was no statistical interaction between translucency vs. handling/modeling techniques and color stability vs. handling/modeling techniques. Adhesive as a modeling liquid protected the composite against sorption and solubility (if powdered gloves were used) and improved its physical/mechanical properties. Translucency and color stability were not correlated with modeling techniques.


Introduction
When the dentist performs composite resin restorations, there are several possibilities of contamination of the operative field that can impair the longevity of the dental restoration [1]. The contamination of the composite resin may negatively affect the clinical results for dental patients. The appropriate use of a rubber dam prevents the contamination of the field and material by the patient's saliva and, eventually, gingival blood. Scientific evidence has shown the increased survival rate of adhesive composite resin restorations when they were performed under rubber dam isolation in relation to the ones that were performed under a cotton roll isolation [2,3]. Nevertheless, due to the high sensitivity of composite resins [4], even using rubber dam isolation, during the manipulation of the material or during its insertion into the dental preparation, there is a chance of contamination and potential negative implications for the restoration. Depending on the physicochemical characteristics of the composite resin, the dentist may choose different methods of manipulation for specific materials. Composite resins that present low flowability are commonly handled and manipulated with gloved hands before insertion into the preparation [5] or using a spatula (or microbrush) covered with a little amount of adhesive to facilitate modeling [6].
A few studies have suggested that the direct contact of gloved hands on the composite resin will contaminate the material due to the powder added on the latex gloves. The contamination with gloves powder can reduce the cohesive strength [7] and microhardness [8] of the composite, consequently jeopardizing the quality and durability of the restoration [7,9]. Adhesive systems used as lubricants to gain flowability may also impair the restoration, creating small defects and propagating pre-existing cracks in between the incremental layers of the composite resin [10].
Despite the available evidence showing some negative consequences of gloves-powder and adhesives on properties of composite resins, the investigation of a more complete assortment of material properties, and the construction of a more clinical-related in vitro experiment, would supply the strongest evidence to help the selection of modeling techniques.
Therefore, considering that the use of a combination of modeling techniques (glovedhands and/or adhesive) may facilitate an appropriate adaptation of the composite resin into the tooth preparation, this study investigated if modeling techniques (combined or not) affected the following physicochemical properties of a composite resin: flexural strength/modulus of elasticity, fracture toughness, sorption/solubility, weight loss, translucency, and color stability.
The null hypothesis was that all the tested properties of a composite resin modeled with gloved-hands and/or adhesive would remain identical to the modeling technique using only a spatula (control group).

Composite Resin Specimens' Preparation
The specimens were prepared using three increment-layers of the composite resin inserted into metal matrices with dimensions appropriated for each physicochemical test: strength/modulus of elasticity (25 × 2 × 2 mm), fracture toughness (25 × 5 × 3 mm), sorption/solubility (15 × 1 mm), and translucency/color stability (10 × 1 mm). After accommodation of the 3 increment-layers into the matrix, the specimen was then lightcured in three different areas (in the center and at both ends) for 20 s for each area with a curing light (Radii-Cal curing light, 1200 mW/cm 2 , SDi, Baywaster, Victoria, Australia).

Experimental Groups
The groups of composite resin specimens were divided according to the modeling/handling techniques used to handle/model the material into the stainless-steel matrices, as follows: (1) Gloved/hands group: each increment in composite resin was handmodeled for 10 s each and inserted into the matrix with a metallic spatula. The operator prepared each specimen wearing a brand-new pair of gloves. (2) Adhesive group: each increment in composite resin was inserted into the matrix with a clean titanium-silicate spatula and modeled using a microbrush (Regular, 2.0 mm, Vigodent, Rio de Janeiro, RJ, Brazil) soaked with adhesive. A brand-new microbrush was used for each specimen preparation. (3) Gloved/hands + Adhesive group: this group is a combination of the previous two groups, i.e., each increment in composite resin was hand-modeled for 10 s each, inserted into the matrix, and modeled using a microbrush soaked with adhesive. (4) Control group: each increment in composite resin was inserted into the matrix with a titanium-silicate spatula (Indusbello Co., Londrina, PR, Brazil). No adhesive and no touch with gloved hands. In between each specimen preparation, the spatula was cleaned with 70% alcohol and dried with absorbent paper.

Flexural Strength and Modulus of Elasticity Analyses
To measure flexural strength and modulus of elasticity of composite resin specimens handled/modeled under different circumstances, n = 10 specimens were prepared for each experimental group. Bipartite stainless-steel matrices (ODEME Biotec. Joaçaba, SC, Brazil) with internal dimensions according to ISO 4949 were used to accommodate the composite resin. The matrix was rested on a polyester strip over a 1 mm thick glass slide. The increment-layers of composite resin were accommodated into the matrix. A second set of polyester strips + glass slide were placed over the matrix/specimen. The specimen was light-cured, removed from the matrix, cleaned from the excesses of composite resin (using a scalpel with blade), and immersed in distilled water into a closed recipient (capped and opaque). Specimens remained stored at 37 • C for 24 h. Prior to testing, specimens were measured with a digital caliper (Mitutoyo Corp, Tokyo, Japan) with an accuracy of 0.001.
The three-point bending test was performed on a universal testing machine (Instron 3342 Single Column, Instron, Canton, MA, USA). The distance between the two supports at its ends was 20 mm and the application of the force in the middle was at a speed of 1 mm/min. The flexural strength was calculated using the following formula: where FS is the flexural strength (MPa), F is the load required for fracture, l is the distance between the supports, and b and h are, respectively, the height and length of the specimen (mm). Data used to obtain the elastic modulus were taken from the straight part of the stress-strain curve in the graph originating from the flexural strength test with digital software (Bluehill, Instron, Canton, MA, USA).

Fracture Toughness Analysis
To measure the fracture toughness of composite resin specimens handled/modeled under different circumstances, n = 6 specimens were prepared for each experimental group. Stainless-steel matrices (ODEME Biotec, Joaçaba, SC, Brazil) with internal dimensions according to ASTM E-699 and a slit of 2.8 mm were used to accommodate the composite resin. Identical procedures performed for the previous tests were executed, such as: inserting incremental layers of composite into a set of polyester strips + thin glass slide, light curing, excess removal, storage, and digital caliper measurement. The fracture toughness was calculated using the following formula: where K Ic = stress intensity factor; P = fracture load; L = distance between the device supports; w = specimen length; b = specimen thickness; a = notch depth.

Sorption and Solubility Analyses
To measure the sorption and solubility of composite resin specimens handled/modeled under different circumstances, n = 5 disk-like specimens were prepared for each experimental group according to the ISO 4049:2008. Identical procedures performed for the previous tests were executed: inserting incremental layers of composite into a set of polyester strips + thin glass slide, light curing, and excess removal. The storage comprised immersing the specimens in distilled water at 37 • C for 30 days. Throughout this period, specimens were removed from the distilled water, dried with absorbent paper, and weighted in an analytical scale (AUW 220D, Shimadzu Corp., Kyoto, Japan) at baseline (day-0), day-7, day-14, day-21, and day-30. These measurements provided the sorption values. Specimens were then dehydrated in a vacuum desiccator over 10 days, and subsequently weighted again to obtain the solubility values.
Sorption and solubility were calculated according to ISO 4049 using the formulas: Sorption = m2-m3/V, where m2 = absorbed mass; mf = final mass; V = volume Solubility = m1-m3/V, where m1 = initial mass; m3 = final mass; V = volume The volume was calculated as a mean of three digital caliper measurements (in different points) of area and thickness around the specimen.

Thermal/Weight Loss Analysis
Thermogravimetric analysis (TGA/DSC) for resin composites was performed on a STA-449-C device (Netzsch, Instruments Inc., Burlington, MA, USA) using a platinum pan as a reference material. Analysis was carried out between 38 • C and 600 • C at the heating rate of 10 • C/min under an inert nitrogen environment with a flow rate of 50 mL/min.

Translucency and Color Stability Analyses
To measure translucency and color stability of composite resin specimens handled/ modeled under different circumstances, n = 6 specimens were prepared for each experimental group. Specimens were prepared using a stainless-steel matrix (calibrated at a 1 mm thickness and a 10 mm central orifice/diameter) (ISO/TR 28642:201151), aiming to create a shade scale (Porcelain Sampler (Smile Line, St-Ilmier, Switzerland)) [11]. A metallic matrix was used to accommodate the incremental layers of composite resin, and the set matrix/composite was gently pressed with a glass slide to guarantee the surface smoothness. Light curing was performed with a Radii Cal (Sdi) curing light for 20 s, according to the manufacturer's specifications. The distance between the light source and the specimen was standardized to 1.0 mm using a glass coverslip. The tip of the curing light came into contact with the glass slide during the photopolymerization process. Specimens were then stored in small recipients and immersed in water at 37 • C for 6 months. The immersion solution was changed every 10 days.
Translucency of the samples was measured using the translucency parameter method, where the color parameters of each specimen were recorded according to the CIE L* a*b* scale on white and black backgrounds. In this system, L* indicates the luminosity in which the average varies from 0 (black) to 100 (white); a*b* indicates the hue; a* represents saturation in the red-green axis; b* represents saturation in the blue-yellow axis. Translucency was calculated using the following formula: where L* W , a* W , and b* W were measured in the white background, and L* B , a* B , and b* B were measured in the black background. Before each measurement, specimens were washed in running water for 10 s and lightly dried with an absorbent paper towel, and the spectrophotometer was calibrated according to the manufacturer's specifications. Specimens were then positioned on a white and black background and were measured 3 times in each background to obtain an average. Color stability was carried out with the color parameters (∆E) obtained in the translucency test on the white background, which were used to calculate the color change after storage in water, in the same experimental period. ∆E was calculated using the formula: where ∆L*, ∆a*, and ∆b* are the difference between the final and initial color parameters L*, a*, and b*, respectively.

Data Analysis
Statistical analysis was conducted using SigmaPlot 13.0 software (Systat Software Inc., San Jose, CA, USA) All data were subjected to the Shapiro-Wilk test to determine data normality. Data originated from the flexural strength, elastic modulus, fracture toughness, and sorption/solubility tests were compared between groups using One-Way (1-factor) and Holm-Sidak ANOVA for contrast of means (α = 0.05). Data originated from the translucency and color stability tests were analyzed using the two-way repeated-measures ANOVA (manipulation vs. time) and Holm-Sidak tests (α = 0.05). Data originated from the thermogravimetric analysis were descriptively reported: percentage of weight loss in relation to the temperature.

Results
Data resulting from flexural strength (MPa), modulus of elasticity (Gpa), and fracture toughness (MPa.m 0.5 ) tests for composite resin specimens handled/modeled under different techniques are shown in Table 2. For these three experimental tests, the highest values were obtained for Adhesive and Gloved/hands + Adhesive (p < 0.05); the lowest values were obtained for Control and Gloved/hands (p < 0.05).
Data resulting from sorption and solubility (µg/mm 3 ) tests are also shown in Table 2. The highest sorption values were obtained for Control. The other groups (Gloved/hands, Adhesive, and Gloved/hands + Adhesive) had similar sorption values (p > 0.05). The highest solubility values were obtained for Gloved/hands, statistically significant (p < 0.05) in relation to the other groups; Adhesive and Gloved/hands + Adhesive had similar solubility values (p > 0.05); the lowest values were obtained for Control (p < 0.05). The thermogravimetric analysis (TGA) showed a similar total mass loss for the groups: Gloved/hands (21.7%); Adhesive (24.0%); Gloved/hands + Adhesive (25.0%); Control (22.8%). Figure 1 shows the TGA curves for composite resin specimens handled/modeled under different techniques. The weight loss for all samples/groups initiated approximately at 300 • C, and the highest weight loss was observed at 450 • C. Table 2. Data (mean ± standard deviation) resulting from flexural strength (MPa), modulus of elasticity (Gpa), and fracture toughness (MPa.m 0.5 ) tests for composite resin specimens handled/modeled under different techniques (gloved-hands and/or adhesive) *. Data resulting from sorption and solubility (µ g/mm 3 ) tests are also shown in Table 2. The highest sorption values were obtained for Control. The other groups (Gloved/hands, Adhesive, and Gloved/hands + Adhesive) had similar sorption values (p > 0.05). The highest solubility values were obtained for Gloved/hands, statistically significant (p < 0.05) in relation to the other groups; Adhesive and Gloved/hands + Adhesive had similar solubility values (p > 0.05); the lowest values were obtained for Control (p < 0.05).

Modulus of Elasticity (Gpa)
The thermogravimetric analysis (TGA) showed a similar total mass loss for the groups: Gloved/hands (21.7%); Adhesive (24.0%); Gloved/hands + Adhesive (25.0%); Control (22.8%). Figure 1 shows the TGA curves for composite resin specimens handled/modeled under different techniques. The weight loss for all samples/groups initiated approximately at 300 °C, and the highest weight loss was observed at 450 °C. Translucency test results are shown in Table 3. There was no statistical interaction between the main factors (p = 0.270). No statistical difference was found for 'handling/modeling technique' (p > 0.05), but a statistical difference was found for 'experimental time' (p < 0.001). Intra-group analysis had the following statistical results for experimental time: baseline/day-0 > 24 h > day-7 > day-90 = day-180 (p > 0.583). Color stability test results are also shown in Tables 4 and 5. Regarding the ∆L parameter, there was no statistical interaction between the main factors (p = 0.355). No statistical difference was found for 'handling/modeling technique' (p > 0.05), but a statistical difference was found for 'experimental time'. Intra-group analysis had the following statistical results: 24 h < day-7 < day-90 < day-180. Regarding the ∆E parameter, there was no statistical interaction between the main factors (p = 0.065). No statistical difference was found for 'handling/modeling technique' (p > 0.05), but a statistical difference was found for 'experimental time'. Intra-group analysis had the following statistical results: 24 h < day-7 < day-90 < day-180.

Discussion
This study considered that the use of a combination of modeling techniques (powdered gloved hands and/or adhesive) may facilitate an appropriate adaptation of the composite Our findings showed that the use of a small amount of hydrophobic adhesive in between the incremental layers of the restoration favored the mechanical properties of the composite resin (flexural strength, modulus of elasticity, and fracture toughness). We also found that the use of adhesive protected the composite resin from sorption and solubility. However, gloves powder negatively affected the solubility of the restoration. Thermal analysis found a higher mass loss for composites modeled with adhesive. In addition to that, this study showed no correlation of handling/modeling techniques with translucency and color stability at the 180-days follow-up. Nevertheless, we did reinforce the findings that composite resin loses translucency and increases color stability over time. These findings rejected our null hypothesis as adhesive and gloved hands interfered in some of the tested composite resin properties.
It is noteworthy to say that the adhesive was the contributor to improve physical properties as the handling/modeling technique only with gloved hands had the same results than the control group (no adhesive and no touch with gloved hands). Previous authors have shown that powdered-gloves impaired mechanical properties of resin composites (powdered-gloves were even more damaging than saliva contamination), suggesting then that latex gloves should be cleaned with ethanol when the hand shaping technique is chosen [7]. In our study, powdered gloves were inert, at least in regard to mechanical properties.
Adhesive made the composite more rigid and prone to elastic deformation before breaking or permanently deforming, and more resistant to failure by cracking. These findings are important because in the dental clinic setting, high values of flexural strength, modulus of elasticity, and fracture toughness are expected from posterior composites to withstand the occlusal forces of mastication and preserve the adhesive interface [12]. Reasons for the improved mechanical properties with the use of a small amount of adhesive between the incremental layers of composite might have been the following: (1) adhesive prevented the incorporation of air bubbles in between adhesive-composite layers, (2) adhesive promoted a better interaction between the composite-composite layers, favoring the composite cohesive strength, which, consequently, reduces the air incorporation and defects into a composite layer [5,11,13].
Up until now, studies investigating the sorption, solubility, mass loss, translucency, and color stability of composites modeled with powdered gloved hands and/or adhesives were absent in the dental literature. Our study was the first to consider this interaction and showed that adhesive modeling protected composites from sorption and solubility. One hypothesis that could explain these findings is that the adhesive used in this experiment had a high concentration of hydrophobic monomers, which acted as a protection barrier and reduced the susceptibility of hydrolyzation of the composite resin compounds [14][15][16]. However, even though the adhesive used in this present study contained a high concentration of hydrophobic monomers, some solubility has been reported for these types of adhesives [14], due to their low rates and degrees of conversion: below 70% [17]. This low degree of conversion may be indirectly observed in our thermal analysis, which showed higher degradation for the composite groups modeled with the adhesive system, possibly due to the presence of HEMA [18]. Additionally, our composite samples that were handled/modeled with gloves presented the highest solubility, and this is a consequence of the powder that is added to the latex, which is extremely soluble and prone to water/humidity degradation [19]. Therefore, if using latex gloves (as those were tested in the present study) for manipulation of the composite, powder-free ones should be preferred. Additionally, evaluations of the influence of nitrile gloves on composites are necessary.
Extremely high solubility values for all experimental groups found in the present study can be explained due to methodological variations, such as: specimen size, storage Translucency and color change are relevant properties for composites mainly when used in esthetic areas. Ideally, both properties should remain the same over time to benefit our patients. However, in general, composites suffer from degradation because of their polymeric nature, which results in a compromised color appearance over time [6]. In the present study, the results did not show statistical differences between the groups tested; a difference was only found between the evaluation times. This means that powdered gloves and adhesive did not affect esthetic properties of the composite resin. Some of the previous articles on this subject showed that the use of adhesive as modeling liquid for composites reduced the alterations in optical properties caused by staining solutions [6,11,20]. These different results may be originated from the differences in methodology amongst the studies, such as: type of composite resins, fabrication of specimens, type of adhesives, immersion media, and evaluation time. Although our study found that the tested esthetic properties were not influenced by powdered gloves and by a hydrophobic adhesive, we did find that all composite samples lost translucency and increased color stability over time, reinforcing the settled results from the literature [21].
This current study has strength and weaknesses. The main limitation of it is the impossibility to directly transport the findings to the clinical setting, as any other laboratorial experiment [22]. In addition, it may be challenging to create a real clinical scenario using different experimental times that are recommended to evaluate several composite properties (mechanical tests were carried out 24 h after specimens' preparation; sorption/solubility tests, after 28 days; esthetic tests, after 180 days). Another limitation is the use of only one type of glove and adhesive. It is important to point out that in the clinical setting, the amount of adhesive and the handling techniques may vary between operators, which could lead to different results. Despite those challenges, our results can guide the decision-making process for choosing a modeling technique for composites, in addition to possibly inspiring further in vivo investigations. The practical relevance of our findings is that dentists may use hydrophobic adhesive as a modeling substance (with or without powdered gloves) when they are performing composite restorations under the incremental layer technique and expect no negative consequences for the material.

Conclusions
In conclusion, the modeling technique using hydrophobic adhesive in between the incremental layers of a composite resin protected the restoration against sorption and solubility (30-days follow-up) and improved the mechanical properties (flexural strength, modulus of elasticity, and fracture toughness) of the composite. The powdered gloved handling/modeling technique did not interfere with the physical/mechanical properties and, also, protected the composite against sorption; however, gloves powder negatively affected the solubility of the restoration. Translucency and color stability (180-day followup) were not correlated with gloved-hands and adhesive modeling techniques.
In sum, our findings, even in light of the limitations of an in vitro study, suggest that modeling techniques using adhesives favor physicochemical properties of composite resins. Finally, adhesives, used in between the incremental layers of the restoration, are not expected to interfere with esthetic properties (translucency and color stability) of the material. Touching composites with powdered latex gloves makes the composite more soluble.