Software SciGlass can be used for determination of processing parameters and to design glass compositions with target properties.
Glasses of the system SiO 2 –B 2 O 3 –Al 2 O 3 –La 2 O 3 –TiO 2 with high refractive indexes and low viscosities can be prepared based on predictions obtained by SciGlass.
Matching the refractive indexes of glass and ceramic preform is crucial to decrease light scattering in ceramic composites.
It is also important to decrease the absorption phenomena to produce composites with high translucency.
The proposed systematic approach was successfully applied to develop glass infiltrated ceramic composite with high light transmittance.
Ceramic composites are promising materials for dental restorations. However, it is difficult to prepare highly translucent composites due to the light scattering that occurs in multiphase ceramics. The objective of this work was to verify the effectiveness of a systematic approach in designing specific glass compositions with target properties in order to prepare glass infiltrated ceramic composites with high translucency.
First it was necessary to calculate from literature data the viscosity of glass at the infiltration temperature using the SciGlass software. Then, a glass composition was designed for targeted viscosity and refractive index. The glass of the system SiO 2 –B 2 O 3 –Al 2 O 3 –La 2 O 3 –TiO 2 prepared by melting the oxide raw materials was spontaneously infiltrated into porous alumina preforms at 1200 °C. The optical properties were evaluated using a refractometer and a spectrophotometer. The absorption and scattering coefficients were calculated using the Kubelka–Munk model.
The light transmittance of prepared composite was significantly higher than a commercial ceramic–glass composite, due to the matching of glass and preform refractive indexes which decreased the scattering, and also to the decrease in absorption coefficient.
The proposed systematic approach was efficient for development of glass infiltrated ceramic composites with high translucency, which benefits include the better aesthetic performance of the final prosthesis.
The most widely used system to produce fixed partial dentures (FPDs) is the metal–ceramic, which is constituted of a metallic framework veneered with a dental porcelain. High clinical success rates for this system have been reported, varying between 72% and 87% after 10 years . However, the opacity of the metal jeopardizes the final aesthetic outcome of the prosthesis, as it is difficult to mimic the translucent dental tissues . Such aesthetic drawback pushed the development of all-ceramic systems, in which the framework can be built with a glass–ceramic, a glass–infiltrated ceramic composite or a polycrystalline ceramic . In comparison to metals, these ceramic materials have excellent biocompatibility, high color stability, and low thermal conductivity . However, it was observed that for most commercial dental ceramics, the higher the mechanical strength, the lower the translucency . Therefore, it is difficult to obtain ceramic frameworks with high mechanical properties and acceptable aesthetic qualities at the same time.
Glass-infiltrated ceramic composites are promising materials that have the potential to fulfill the requirements of having both high toughness and high translucency. Nevertheless, unfortunately the commercial glass–infiltrated composites currently available in the dental market do not fulfill these requirements . These materials still have low translucency due to an important mismatch of the refractive indexes of the different microstructural phases that results in light scattering at the interfaces, decreasing light transmittance through the body . Indeed, ordinary silicate glasses usually have low refractive indexes ( n = ∼1.5) compared to most reinforcing oxide phases, like alumina, spinel (MgAl 2 O 4 ) and ceria-stabilized zirconia ( n = 1.7–2.2). The addition of titanium oxide and lead oxide can significantly increase the refractive index of silicate glass and help solving the light scattering problem in these composites , but because of its toxicity, it is not safe to use lead oxide in biomaterials. Some aluminosilicate glasses containing rare earth oxides have relatively high refractive indexes ( n = 1.7–1.8), especially after the addition of lanthanum and yttrium oxides . This explains why glasses that are infiltrated into alumina preforms usually contain lanthanum oxide . Another important glass property is the viscosity, since it affects the temperature at which the infiltration process is carried out. However, the literature lacks information about the viscosity values of glasses in the range of the infiltration temperature.
The aim of this study was to verify the effectiveness of a systematic approach using a software that takes into account the glass properties to predict the processing parameters related to glass infiltration. This work also investigated the capability of this systematic approach in designing specific glass compositions with target properties in order to prepare glass infiltrated ceramic composites with high translucency.
Materials and methods
In this work, alumina (Al 2 O 3 ) was chosen as the ceramic phase for the composite preform because of its good mechanical properties (421 GPa of Young’s modulus, 4.0 MPa m 1/2 of fracture toughness, 397 MPa of four-point flexural strength, and 19.8 GPa of Vickers hardness), higher than spinel (MgAl 2 O 4 ) ceramic . Although the preform of zirconia–toughened alumina results in a stronger composite , it is difficult to match the refractive index of the infiltrating glass to both ceramic phases, since they have significantly different refractive indexes (1.76 and 2.19 for alumina and zirconia, respectively ).
Simulation of glass properties and compositional design
The software SciGlass 7.7 (Glass Property Information System, Lhasa) was used to calculate the viscosity and refractive index of the glass from its chemical composition. In order to design the glass compositions, it was necessary to know the viscosity of the glass at the infiltration temperature, in order to restrict the maximum temperature to 1200 °C, which is the capacity of commercial furnaces for this process. Therefore, it was necessary to calculate this viscosity based on the data provided by the literature. Fig. 1 shows the calculated viscosities from the chemistry of some glasses as a function of the temperature used for the infiltration process reported in Refs. . It could be observed that the viscosities of the glasses at the infiltration temperature ranged from around 10 2 to 10 3 dPa s. Therefore, in this work it was established that the glass should have viscosity lower than 10 2 dPa s at 1200 °C for the design of glass composition.
The basic glass composition chosen was 45% SiO 2 , 25% Al 2 O 3 , 15% La 2 O 3 , and 15% TiO 2 (mol%), since a similar composition was reported to result in glass with refractive index close to that of alumina. The simulation of this composition with software SciGlass 7.7 indicated viscosity of 10 2 dPa s at 1422 °C for this glass, but this temperature was too high for the scope of this work. In this way, the addition of alkaline oxides (Li 2 O, Na 2 O, K 2 O, Rb 2 O, and Cs 2 O) was simulated in order to replace silica in this glass, however instead of decreasing the temperature, this change resulted in an increase in the temperature at the viscosity of 10 2 dPa s ( Fig. 2 a ). The simulations showed temperature dependence of viscosity that was similar for all alkaline oxides up to the addition of 25 mol% ( Fig. 2 b). The addition of most of the alkaline-earth oxides (MgO, CaO, SrO, and BaO) to the basic glass composition also increased the temperature at the viscosity of 10 2 dPa s ( Fig. 2 c), except for the addition of beryllium oxide which lowered this temperature to 1319 °C for 25 mol% addition ( Fig. 2 d), but not sufficiently to the target temperature (<1200 °C).
The simulations, however, indicated that the addition of boron oxide could significantly lower the temperature at the viscosity of 10 2 dPa s of the glass 45SiO 2 –25Al 2 O 3 –15La 2 O 3 –15TiO 2 ( Fig. 3 ). The addition of 15 mol% or higher amounts of B 2 O 3 replacing SiO 2 could achieve the target temperature (<1200 °C). In this work the glass composition chosen for the experimental section was 20SiO 2 –25B 2 O 3 –25Al 2 O 3 –15La 2 O 3 –15TiO 2 (mol%), since the simulations indicated adequate viscosity (10 2 dPa s at 1115 °C) for the infiltration process and refractive index of 1.764, close to the alumina ( n = 1.76 ).
Preparation of glass by melting
The following particulate raw materials were used: quartz sand (>99.5% SiO 2 , Jundu); boric acid (>99.5% H 3 BO 3 , PA ACS, Vetec); Al 2 O 3 (99.9%, UA5105, Showa Denko); La 2 O 3 (99.9%, PA, Vetec); and TiO 2 (>99%, 1001, Kronos). The raw materials were weighted to prepare a mixture of 100 g to achieve the following composition: 20% SiO 2 , 25% B 2 O 3 , 25% Al 2 O 3 , 15% La 2 O 3 , and 15% TiO 2 (mol%), considering the theoretical mass loss of water from the boric acid.
After manual mixing, the powder mixture was melted in a 100 mL Pt–5%Au crucible at 1500 °C for 1 h using an electrical furnace (FE-1700, Fortelab). Then, part of the melt was poured into the distilled water to prepare the frit and the remaining melt in a graphite mold to prepare the disc-shaped samples (diameter of 20 mm). The frit was milled in a ball mill with plastic jar and alumina balls to prepare a powder finer than 100 mesh (150 μm). The discs were annealed in an electric muffle (1612, Jung) at 810 °C (above the glass transition temperature, T g = 740 °C, calculated using SciGlass 7.7) for 1 h and cooled slowly at the rate of 0.5 °C/min to relieve the internal residual stresses.
Preparation of the composite by infiltration
The alumina preforms were prepared from commercial porous presintered alumina blocks with 25.5% porosity (CA-12, In-Ceram Alumina, Vita Zahnfabrik) by cutting with diamond saw aiming at dimensions of 10.5 × 12.5 × 1.5 mm 3 .
The glass powder mixed with a small amount of water was placed on the surface of the porous alumina preforms and afterwards taken into an electric furnace (Kerampress, Kota). The thermal cycle for spontaneous infiltration was carried out under vacuum with heating rate of 60 °C/min up to 1200 °C and maintained at this temperature for 60 min; then, the furnace was switched-off.
Characterization of glass and composite
The prepared glass powder was analyzed by X-ray diffraction (XRD) analysis using a diffractometer (D8 Focus, Bruker) and Cu-Kα radiation with scan step of 0.05° and 2 s of counting per step in order to confirm its amorphous structure. The refractive index of the glass was determined on the disc sample with polished surfaces using a refractometer (2010/M Prism Coupler, Metricon) at the wavelength of 633 nm.
The microstructure of the prepared composite was analyzed using a scanning electron microscope (SEM, Quanta 600 FEG, FEI) coupled to an energy dispersive spectrometer (EDS, Xflash Quanta 400, Bruker). The crystalline phases were identified by XRD analysis.
The composite samples with polished surfaces and thickness of 1.0 mm were analyzed in a spectrophotometer (CM-3700d, Konica Minolta) to measure the total transmittance and the diffuse reflectance, using white and black backgrounds with glycerol as coupling liquid, in the visible light region (wavelength from 360 to 740 nm) with 10 nm measurement interval. For comparison, samples of a commercial glass infiltrated alumina composite (InCeram Alumina, Vita Zahnfabrik), prepared with the same porous alumina preform used in this study and the commercial glass powder (AL 2 In-Ceram Alumina Glass Powder, Vita Zahnfabrik), and a commercial lithium disilicate reinforced glass-ceramic (IPS Emax Press, Ivoclar Vivadent), prepared by the heat-pressing technique, with polished surfaces and thickness of 1.0 mm were also analyzed. The contrast ratio, which is a measure of the opacity of a material, was determined by the ratio of the reflectance values with black and white backgrounds . The scattering ( S ) and absorption ( K ) coefficients were determined by Kubelka–Munk (K–M) model using the following equations :
r d = 1 2 + ( n − 1 ) ( 3 n + 1 ) 6 ( n + 1 ) 2 + n 2 ( n 2 − 1 ) 2 ( n 2 + 1 ) ln n − 1 n + 1 − 2 n 3 ( n 2 + 2 n − 1 ) ( n 2 + 1 ) ( n 4 − 1 ) + 8 n 4 ( n 4 + 1 ) ( n 2 + 1 ) ( n 4 − 1 ) 2 ln ( n )
a = 1 2 R sw * R w * R b − R sw * R w * R sb * + R b − R sb * − R w * R sb * R b − R w + R sw * R sb * R b + R sw * R b R sw * − R w * R sb *