Evaluation of 5Y-PSZs with the same phase composition but different microstructures and crystal structures of tetragonal phase.
Density, hardness and toughness were comparable.
Translucency, strength and aging-stability were considerably different.
The advantages of PSZ ceramics need to benefit from both c – and t -ZrO 2 phases.
For the use of partially stabilized zirconia (PSZ) as high-translucent zirconia, the importance of cubic phase ( c -ZrO 2 ) is commonly emphasized without much attention for the remaining tetragonal phase ( t -ZrO 2 ). The aim is to understand whether the crystal structure and microstructure of t -ZrO 2 have paramount influences on the properties of PSZ.
Two grades of 5 mol% yttria-stabilized PSZ ceramics were prepared by different processing routes. A 5 mol % yttria co-precipitated zirconia (T5Y) was compared with a 3 and 8 mol% yttria-stabilized powder mixture grade with a bimodal grain size distribution (B5Y). The phase composition and lattice parameters were studied with x-ray diffraction (XRD) and Rietveld analysis. Mechanical properties were compared in terms of hardness, toughness and biaxial bending strength with Weibull analysis. Their translucency and aging stability were also characterized, and the microstructure before and after aging were followed by scanning electron microscopy (SEM).
T5Y and B5Y having similar phase composition (about 40% t – and 60% c -ZrO 2 ) showed the same basic properties (density, hardness and toughness), but the lattice parameters of t -ZrO 2 and the concomitant microstructure were completely different. The t -ZrO 2 phase in T5Y had a higher yttria content and a lower tetragonality (c/a) of 1.0126 ± 0.0002, whereas the t -ZrO 2 phase in B5Y was comparable to that in conventional 3Y-TZPs with a tetragonality of 1.0153 ± 0.0002. Consequently, B5Y showed a lower strength with lower Weibull modulus, being less translucent and more aging susceptible as compared to T5Y. The translucency and aging susceptibility of B5Y were even comparable to those of conventional 3Y-TZP.
Not only the phase composition with high cubic phase content but also the lattice parameters and the yttria content in the remaining t -ZrO 2 played a determinant role for PSZs to be used as ‘high-translucent’ zirconia.
Driven by biocompatibility concerns and increased esthetic requirements, there is a clear trend towards metal-free dental restorations, of which all-ceramic restorations are very popular [ , ]. With the advance of CAD/CAM, yttria-stabilized zirconia ceramics have been used as dental restorative materials for about two decades [ ]. There are several variants of yttria-stabilized zirconia ceramics, depending on the sintering conditions, additives and especially the yttria stabilizer content [ , ]. The conventional and the most widely used is 3 mol% yttria stabilized tetragonal zirconia (3Y-TZP), with exceptional mechanical properties – the highest strength (ability to prevent crack initiation) and toughness (ability to provide damage tolerance once cracks form) ever reported for dental ceramics [ ] – benefiting from transformation toughening [ , ]. However, this zirconia is opaque and is typically used as the framework for fixed dental prostheses (FDPs). More translucent 3Y-TZP ceramics (i.e. second-generation dental zirconia) were later obtained by limiting the alumina content that acts as light-scattering defects [ , , ]. Nevertheless, in general 3Y-TZP ceramics still allow less light transmission than glasses or glass-ceramics [ ], primarily because of the large birefringence effect of tetragonal phase zirconia ( t -ZrO 2 ) [ ]. Tetragonal zirconia and in fact all non-cubic crystalline ceramics are optically anisotropic (i.e. birefringent), implying that the refraction index is anisotropic in different crystallographic directions, causing light scattering including reflection and refraction at grain boundaries [ , ]. In order to reduce the grain-boundary birefringence in zirconia ceramics, partially stabilized zirconia (PSZ) containing a higher yttria content like 4 or 5 mol%, have very recently been introduced in the dental field. This strategy has been approved to be a robust way to enhance the translucency of zirconia ceramics [ , , , , ] and considered as third-generation dental zirconia [ , , ].
In addition, 3Y-TZP ceramics are known to be susceptible to hydrothermal aging, during which spontaneous phase transformation from tetragonal ( t ) to monoclinic ( m ) zirconia takes place in a humid environment at moderate temperature including human body conditions [ ]. Although to date no clinical failure of dental restorative zirconia is reported due to its hydrothermal susceptibility [ ], aging generally results in surface roughening and subsurface microcracks [ ], worsening the zirconia characteristics. In severe cases, aging may decrease zirconia’s mechanical [ ] and optical properties [ ]. In this light, zirconia stabilized by a higher yttria content normally have a higher aging stability [ , ], lowering the associated risks.
However, an increased yttria content reduces or even prevents the transformation toughening effect and mechanical performance in terms of flexural strength, toughness, reliability and fatigue strength [ , , , ], by which these so-called ‘high-translucent’ third-generation zirconia ceramics are inferior to 3Y-TZP ceramics. High strength 3Y-TZP can be used for single crowns, short to long-span bridges in load-bearing areas [ ] and recently also for implants [ , ]. Yet, the flexural strength and toughness of 5 mol% yttria-stabilized zirconia can drop by a factor of about 2, approaching the values recorded for glass ceramics [ ], by which 5Y-TZP is mainly indicated for anterior crowns.
Regarding the underlying mechanism as to why an increased yttria content can give rise to zirconia ceramics with higher translucency, better aging-stability and lower mechanical properties, the vital change in phase composition is well known [ , , ]. As compared to 3Y-TZP that typically consists of ∼80 wt% t -ZrO 2 and ∼20 wt% c -ZrO 2 , PSZ stabilized with 4−6 mol% yttria used for dental purposes contains 40−70 wt% c -ZrO 2 depending on the yttria stabilizer concentration and sintering temperature [ , , , ]. The importance of introducing more cubic phase zirconia ( c -ZrO 2 ) is well recognized [ , , , , , , ], i.e. cubic zirconia is non-birefringent [ , ] and non-transformable under stresses and water conditions [ , , ]. PSZs (mixture of t -ZrO 2 and c -ZrO 2 ) sometimes are even referred to as ‘fully stabilized zirconia’ [ ] or ‘cubic’ zirconia [ ]. However, not much attention was paid to the remaining t -ZrO 2 and its role in PSZ was not clearly clarified [ , ]. Therefore, the purpose of this study was to investigate whether the nature of the t -ZrO 2 phase has a significant influence on the performances of PSZ ceramics as dental restorative materials. Two PSZ ceramics with the same overall yttria stabilizer content and phase composition were prepared by two different routes in order to obtain two ceramics with a distinct t -ZrO 2 phase structure. The null hypothesis tested was that PSZ ceramics having the same cubic phase composition would show the same mechanical properties, aging stability and translucency, irrespective of the microstructure and crystal structure of the remaining tetragonal zirconia.
Materials and methods
Two partially stabilized zirconia (PSZ) with 5 mol% yttria were prepared from different starting powders. A ‘T5Y’ grade was made from a commercially available powder (ZpexSmile, Tosoh, Tokyo, Japan), for which a co-precipitation method was used to synthesize the powder and the yttria stabilizer was homogeneously distributed in the zirconia grains [ ]. Another grade, ‘B5Y’, was made from a 3 and 8 mol% yttria-stabilized zirconia powder mixture. Appropriate amounts of spray-dried 3 mol% yttria (TZ-3Y, Tosoh) and 8 mol% yttria (TZ-8Y, Tosoh) powders were dry-mixed without milling balls on a multidirectional mixer (Turbula, Basel, Switzerland) at 75 rpm for 24 h. Co-precipitated powders containing 3 and 4 mol% yttria (Zpex and Zpex4, Tosoh), referred to as ‘T3Y’ and ‘T4Y’, were used as reference materials for translucency and aging stability analysis.
The powders were uniaxially pressed at 200 MPa for 40 s (Nannetti SSN/EA, Faenza, Italy) into 18-mm diameter and 4-mm thick disks and cold isostatically pressed (EPSI, Temse, Belgium) at 300 MPa for 3 min, forming ‘green bodies’. All green bodies were debinded, pre-sintered at 1000 °C for 1 h and finally pressureless sintered in air at 1450 °C for 2 h. The sintered ceramics (16-mm diameter) were parallel ground and sequentially polished with 15, 6, 3, 1 μm diamond suspensions to a thickness of 1.2 mm for the characterization of phase composition, microstructure, mechanical properties and hydrothermal aging stability, and to a thickness of 0.5 mm for translucency measurements.
The densification behavior of the ceramics was investigated by dilatometry analyses (DIL 402C/7 Netzsch, Selb, Germany) of pre-sintered materials (1 h at 1000 °C). The change of specimen length (L) was recorded during heating and cooling between 20 °C and 1500 °C at a rate of 5 °C/min. The normalized specimen length, i.e. specimen-length change (ΔL) divided by the initial specimen length (L 0 ) was plotted as a function of temperature. In order to clearly observe the onset temperature of sintering, the first derivative of ΔL/L 0 was also reported as a function of temperature.
The density of the green compacts was measured geometrically and the density of the sintered ceramics was measured according to the Archimedes principle in ethanol.
X-Ray Diffraction (XRD; 3003-TT, Seifert, Ahrensburg, Germany) was used to characterize the phase composition of the sintered ceramics. XRD patterns were collected on polished ceramic surfaces ( n = 6) using Cu-K α radiation at 40 kV and 40 mA from 20 to 90° (2θ) with a step size of 0.01 for 3 s. Rietveld analysis was carefully performed with Topas academic software (Bruker AXS, Karlsruhe, Germany) to quantify the phase contents and lattice parameters. The yttria (Y 2 O 3 ) content in the t -ZrO 2 phase was calculated based on its a and c unit cell parameters, according to the formula [ , ]:
The microstructure was examined by scanning electron microscopy (SEM; FEI-Nova Nanosem 450, FEI, Eindhoven, The Netherlands) of as-polished specimens. Additional surface treatments (thermal etching and coating of conductive layer) were avoided in order to observe the intrinsic microstructures. The grain size distribution of >1000 grains were measured on SEM micrographs according to the linear intercept method without applying a correction factor.
Biaxial bending strength ( n = 20) was measured at a loading rate of 0.5 mm/min with a piston-on-three balls testing set up following the modified ISO 6872 standard [ ]. The diameter of the piston was 1.6 mm and the three balls were placed equidistantly on a circle with a diameter of 12 mm. Instead of polishing the ceramic specimen up to 15−20 μm diamond grit, as suggested in ISO 6872, mirror-polished specimens (up to 1-μm diamond suspension) were used to avoid potential influence of surface stress generated by the rough grinding and polishing processes [ ].
Indentation fracture toughness and hardness was measured using a Vickers micro-hardness tester (Model FV-700, Future-Tech, Tokyo, Japan) with a load of 10 kg ( n = 10) for 10 s. The toughness was calculated from the surface length of the cracks developing at the corners of the Vickers indents according to the Anstis equation [ ] with an E-modulus of 210 GPa.
Translucency was measured on 0.5-mm thick ( n = 6) polished specimens using a spectrophotometer (SpectroShadeTMMICRO, MHT OpticResearch, Niederhasli, Switzerland). The CIELAB coordinates ( L* , a* and b* ) were recorded against white and black calibrated background boards with a thin layer of vaseline in between. The contrast ratio (CR), defined as the ratio of illuminance ( Y ) of the specimen placed on a black ( Y b ) and a white ( Y w ) background, was calculated as Y b / Y w . The value of Y was calculated with the L* value measured by the spectrophotometer, according to [ ]:
with Y n equal to 100 for simulated object colors. A lower CR value indicates a higher translucency and a value of 0.0 implies a transparent material, whereas 1.0 indicates a totally opaque material.
The hydrothermal aging stability was evaluated by accelerated artificial aging. Double-side polished disks ( n = 6) were hydrothermally treated in an autoclave in steam (2 bar) at 134 °C. The monoclinic phase content ( V m ) on the surface was measured by XRD in the range of 27–33° 2θ and calculated according to the formula of Garvie et al. [ ] that was modified by Toraya et al. [ ]: