Composite Resins

The resin matrix for the experimental composites was prepared by mixing bisphenol-A-glycidyldimethacrylate (Bis-GMA, Merck, Darmstadt, Germany) and triethylene glycol dimethacrylate (TEGDMA, Merck) in a weight ratio of 60:40. The resin mixture was rendered photocurable by the addition of 0.2 wt% of camphorquinone (Merck) and 0.8 wt% of ethyl-4-(dimethylamino) benzoate (Merck). All components were mixed using a magnetic stirrer for 48 h.
BG 45S5, inert barium glass, and silica were obtained from commercial vendors. The experimental BG was prepared on-demand by the company Schott (Mainz, Germany) via the melt–quench route. The preparation and grinding procedures for the experimental BG were similar as for BG 45S5 in order to obtain similar particle sizes of both BG types. The experimental BG featured a lower Na₂O content than conventional BG 45S5 (10.5 wt% vs. 24.5 wt%), and additionally contained 12 wt% of CaF₂. The theoretical network connectivity of the experimental BG (2.1) was similar to that of conventional BG 45S5 [9 (link)]. Reinforcing fillers (inert barium glass and silica) were silanized, whereas the BG fillers were used without surface silanization.
Experimental composites were prepared by admixing varying ratios of bioactive and reinforcing fillers (Table 1) into the resin matrix. The series of composites containing 5–40 wt% of conventional BG 45S5 was denoted as the C-series, while the composite series functionalized with the same wt% of the experimental fluoride-containing BG was denoted as the E-series (Table 2). The control composite contained only reinforcing fillers. The total filler load in all composites was 70 wt%. The ratios of BG and reinforcing fillers followed previous studies of experimental BG-functionalized composites [23 (link),24 (link),28 (link),29 (link)].
The resin system and the fillers were mixed using a dual asymmetric centrifugal mixing system (SpeedMixer DAC 150.1 FVZ, Hauschild and Co. KG, Hamm, Germany) at 2000 rpm. Mixing was performed in five one-minute intervals separated by one-minute breaks. After mixing, the prepared composites were deaerated in a vacuum for 48 h.
Three commercial acid-neutralizing materials were used as references, namely, a reinforced glass ionomer restorative (ChemFil Rock, Dentsply Sirona, Konstanz, Germany; shade: A2, LOT: 1807000740), a giomer (Beautifil II, Shofu, Kyoto, Japan; shade: A2, LOT: 041923), and a resin-based “alkasite” material (Cention, Ivoclar Vivadent, Schaan, Liechtenstein; shade: universal, LOT: XL7102). The alkasite material contained two types of reactive filers; an ionomer glass based on a calcium barium alumino-fluoro-silicate, and a calcium fluoro-silicate glass [27 (link)].

Three direct composite resins currently indicated for esthetic anterior and/or posterior
restorations were used in the present study. Information regarding composite type,
composition, curing time and manufacturer is given in Table 1.
Fifteen specimens (15 mm diameter x 2 mm thick) of each composite were fabricated using
a stainless steel matrix. each material was inserted into the matrix in 1.0-mm-thick
increments photoactivated with a halogen light-curing unit (Ultralux, Dabi Atlante,
Ribeirão Preto, SP, Brazil), according to the manufacturer’s information (Table 1). The specimens were polished with aluminum
oxide discs (Sof-Lex, 3M eSPe, St. Paul, MN, USA) in a sequence of decreasing
abrasiveness with intermittent movements, and the specimen surface was kept moist at
each disc change. The polished specimens had their thickness measured with an electronic
digital caliper accurate to 0.1 mm (Digimess, São Paulo, SP, Brazil). The
polished specimens were stored in the dark at 100% of humidity for 24 h.
Color was measured according to the CIe (Commission Internationale de l´eclairage)
L*a*b* system relative to CIe standard illuminant D65, against a white background
(Standard for 45/0 degrees; Gardner Laboratory, Inc, Bethesda, MD, USA) in a reflection
spectrophotometer (PCB 6807 BYK Gardner, Geretsried, Germany). This equipment is
specific for color measurement and has 30 LED lamps with 10 different colors arranged in
a circle, which directs a light bundle at 45º with the material surface. This
light bundle is reflected 0º back to the equipment, which captures and records
the L*, a* and b* values of each specimen. The axis L* refers to the lightness
coordinate and its value ranges from zero (black) to 100 (white). The axes a* and b* are
chromaticity coordinates in the red-green axis and the yellow-blue axis, respectively.
Positive a* values indicate a shift to red and negative values indicate a shift to
green. Similarly, positive b* values indicate the yellow color range and negative values
indicate the blue color range.
After baseline color measurement, the specimens were assigned to three groups (n=5),
each one immersed in a different solution, and subjected to a new color measurement.
Group 1 (control) was immersed in distilled water; Group 2 was immersed in coffee
prepared by dissolving 1.5 g of soluble coffee (Nescafé Classic; Nestlé
SA, Vevey, Switzerland) in 50 mL of boiling distilled water, and Group 3 was immersed in
a cola soft drink (Coca-Cola^®, Refrescos Ipiranga, Ribeirão
Preto, SP, Brazil). The solutions were replaced every day.
After 15 days immersed in the solutions, the specimens were rinsed with distilled water
for 5 min and blotted dry with absorbent paper before the second color measurement.
Color of the specimens was measured after immersion in the different solutions by the
spectrophotometer, as previously described. Using the values of L*, a*, b*, color
variation (De1) between 15-day immersion and baseline measurements was determined using
the following equation:
ΔE1 = [ (ΔL*) ² + (Δa*) ² +
(Δb*)²]½
The values of ∆ > 1 are considered as visually perceptible and values of DE
≥ 3.3 are considered as clinically unacceptable^{18 (link)}. After determination of color variation (De1), the
specimens were repolished with Sof-Lex discs (3M ESPE) as previously described in the
first polishing procedure. The repolished specimens were submitted to a new color
measurement (CIe L*a*b*) and these values were compared to those of the baseline color
measurements, so a second value of color variation (∆2) was obtained.
The means and standard deviations of color change (∆1 and ∆2) were
calculated and submitted to statistical analysis by 3-way repeated measure ANOVA
(variation factors: composite, treatment and evaluation period) and Tukey’s test at 5%
significance level.

Etching with 5% hydrofluoric acid (IPS Ceramic Etching Gel, Ivoclar Vivadent, Schaan, Liechtenstein) for 20 seconds was done on the tissue surfaces of endocrowns in IPS e.max CAD and Vita Suprinity groups and 60 seconds for Vita Enamic. After etching, each restoration was cleaned in an ultrasonic apparatus for five minutes and then dried with oil-free air spray. A thin layer of silane coupling agent (Prosil; FGM) was applied to the internal walls of the endocrowns for 60 seconds and then air-dried.
Self-adhesive resin composite cement (RelyX Unic-em 2 Automix, 3 M ESPE, Seefeld, Germany) with a 1: 1 base-catalyst ratio was mixed to obtain a uniform consistency. The cement was used on the tissue surface of the endocrowns. The restoration was placed on the tooth with a 3 kg weight in a load applicator. The excess cement was removed after 2-3 minutes from the start of the mix. Then the cement was light-activated for 20 seconds. A light-emitting diode curing unit (Demetron A.1, Kerr/Sybron, Orange, CA, USA) with a 12-mm diameter curing light tip
and irradiance output of 1000±50mW/cm² was used. The surface-tip distance was 0.5mm (Figure 5).
After cementation, all samples were kept in an incubator (Model 2; Precision Scientific Co., Columbus, OH, USA) at 37°C for 24 hours.

Pre-cured composite resin cylinders (Filtek Z-250, 3M ESPE, St. Paul, MN, USA) were fabricated in transparent plastic tubes with an inner diameter of 3 mm and a height of 3 mm [12 (link)] using the incremental technique. Each layer of 1 mm thickness was light-cured with a light-curing unit with an intensity output of 1200 mW/cm² (Elipar^TM S10, 3M ESPE, St. Paul, MN, USA). The composite resin cylinders were polished with 600-grit silicon carbide papers (Buehler, USA) under water coolant and cleansed ultrasonically in distilled water for 5 min before use.
The 3Y-TZP disks (N = 64) were randomly divided into two groups (n = 32), and the LDGC disks (n = 32) served as positive control. The ceramic disks were processed and bonded according to the following surface treatments:
(1) Group APA+MDP: 3Y-TZP disks were pretreated with APA, as described in 2.1, ultrasonically cleaned with 99.5% ethanol for 3 min then totally dried with oil-free air spray before being bonded with resin cylinders using MDP–containing resin cement (Clearfil SA Luting cement, Kuraray Noritake Dental, Japan) by mixing equal amounts of paste A and paste B of the cement for 10 s. The bonded specimens were kept under a 5N load [31 (link)] for 3 min [32 (link)].
(2) Group GCSD: 3Y-TZP disks pretreated with GCSD were etched with 5% HF (IPS Ceramic Etching Gel, Ivoclar Vivadent IPS, Liechtenstein) for 90 s, thoroughly water-sprayed for 2 minutes, cleansed and then dried with oil-free air spray as mentioned above. They were then applied with silane agent (Monobond N, Ivoclar Vivadent IPS, Liechtenstein), left undisturbed for 60 s and strongly air-dried for 5 s. Finally, they were bonded with resin cylinders using resin cement without MDP (Variolink N, Ivoclar-Vivadent IPS, Liechtenstein) by mixing equal amounts of base and catalyst paste for 10 s. The subsequent bonding procedure was same as that of Group APA+MDP.
(3) Group LDGC: LDGC disks were etched by HF, applied with silane agent and bonded with Variolink N following the procedure of Group GCSD.
A schematic illustration of the materials and methods for bond strength testing is shown in Figure 1.