G. Heydeckea, b, , , F. Butza, J.R. Binderc and J.R. Struba
aDepartment of Prosthodontics, University Hospital, Freiburg, Germany
bFaculty of Dentistry, McGill University, Montreal, Canada
cInstitute for Materials Research III, Forschungszentrum Karlsruhe, Karlsruhe, Germany
Received 15 August 2005; revised 21 March 2006; accepted 22 June 2006. Available online 1 September 2006.
To confirm the clinical applicability of a novel ZrSiO4 (zircon) based shrinkage-free ceramic material, the flexural strength, fracture toughness and chemical solubility were tested. In addition, the fracture load of full crowns made from this material was tested after cyclic thermomechanical loading.
Flexural strength of 12 specimens was measured using a biaxial bending test. Fracture toughness was measured using 10 slotted box shaped specimens. The specimens were fractured using a universal testing machine; fracture loads were recorded. A chemical solubility test was performed in accordance with ISO norm 6872. Additionally, 32 ZrSiO4 all-ceramic crowns were fabricated on extracted caries-free human molars. Sixteen Empress 2 and 16 PFM crowns served as a reference control. After artificial aging of 1.2 million cycles in the chewing simulator, the survival rate of the crowns was determined. The fracture load of all surviving specimens was obtained by loading the crowns until fracture in a universal testing machine.
A flexural strength of 328 MPa was found. The fracture toughness of the ZrSiO4 ceramic was 5.16 MPa m0.5. The chemical solubility amounted to 7.2 μg/cm2. All specimens survived the chewing simulation (survival rate: 100%); no crowns had to be re-cemented. A mean fracture strength of 1790 N was found for Everest HPC for Empress 2 crowns, 1715 N for Empress 2 crowns and 2416 N for PFM crowns. Fracture loads of PFM crowns were significantly higher than for Empress 2 crowns (P = 0.032) as well as ZrSiO4-crowns (P = 0.007). There was no significant difference between ZrSiO4-crowns and Empress 2 crowns (P = 0.743).
At the present stage, Everest HPC can be recommended for the fabrication of single crowns as an alternative to conventional PFM and other all-ceramic crowns, because its fracture strength exceeds average masticatory forces in the posterior region.
Masticatory efficiency, phonetics and esthetic appearance are essential determinants of patient satisfaction with dental prostheses . However, the individual evaluation of esthetics is very subjective . Over the past decades, patient expectations of dental esthetics have been increasing continuously, particularly as they relate to the selection of tooth color . A high degree of sophistication has been achieved in the realm of tooth color esthetics. Three factors can be held responsible for this development: (1) the influence of highly skilled dental technicians, (2) the continuous refinement of manufacturing techniques, and (3) improvement of dental materials ,  and . However, esthetic alternatives to the full gold crown such as porcelain-fused-to-metal (PFM) crowns are generally more expensive. All-ceramic crowns are even more resource intensive with regard to their manufacture and cost. Social insurance systems mostly cover the least costly option – making the cast gold crown the standard of care.
In addition to the PFM technique, there are quite a number of different systems available for the manufacture of all-ceramic restorations. However, all of those systems were introduced with the aim of further improving the appearance of tooth colored fixed prosthodontics and to extend their indication to the posterior area. These can be considered top end restorations. There are few efforts to improve esthetics at the standard level. CAD/CAM techniques appear to the most suitable for the creation of cost effective fixed restorations. But, so far, they have not been explored to their full potential to provide simple and low cost single crowns. If asking patients, most will answer that natural teeth are “white”. While this may seem a coarse definition to a dental professional this could be a starting point for simple alternatives to a cast gold crown which is not white.
Several techniques have been described for the fabrication of all-ceramic fixed restorations. For example, with aluminium oxide glass infiltrated core ceramics (In-Ceram alumina, Vita, Germany) a flexural strength of >450 MPa can be achieved. With heat pressed lithium–disilicate cores, flexural strengths of slightly less than 400 MPa have been reported . Both types of materials required highly skilled manual techniques for their respective slip-casting, waxing or heat pressing procedures. One of the drawbacks of all sintered ceramics is their linear sinter shrinkage of up to 20%. To obtain the desired dimensions, restorations must be formed to correspondingly enlarged dimensions.
More recent fabrication techniques involve the machining of CAD-designed restorations (e.g., Cerec, Sirona, Germany; DCS, Switzerland; Cercon; Degudent, Germany). The most prevalent core material is Y-TZP zirconia. The machining of restorations from sintered ZrO2is quite time consuming; the machining process for a single unit restoration often takes 2–3 h . In addition, the extreme hardness of sintered high strength ceramics results in high wear of the machining tools. Recent evidence shows that the machining process also has a detrimental effect on the microstructure of the material. As a consequence, the induced microcracks reduce the reliability of restorations machined from fully sintered ZrO2 and . Restorations made using these techniques mostly require a veneering layer to (a) achieve a proper tooth shape and (b) to cover the core material to achieve acceptable esthetics.
An alternative to slip-casting or grinding of sintered blocks is CAD/CAM machining of pre-sintered blocks or blanks at the green stage. The use of a conventional ceramic material process requires an enlarged green body to compensate for the linear sinter shrinkage. The enlargement factor depends on the density of the green compact, the density of the body after sintering and their respective weights. Thus, the precision of the white body directly depends on a constant quality of the green bodies and the stability of the sintering process.
These problem areas can be circumvented if a shrinkage-free ceramic material is used. With a so called reaction bonding process, this is achieved through the volume expansion of one component during the sintering process compensating the volume loss of another. If a low loss binder is added, the sinter shrinkage can be reduced further. Such a system based on zirconium silicide (ZrSi2), zirconia (ZrO2) and a polymethylsilsesquioxane (PMSS) has been described in detail elsewhere . The ceramic based on this system is zircon (ZrSiO4). It can be molded to blanks using axial or isostatic pressing. The blanks are stable and can be machined using commercial CAD/CAM equipment.
2. to assess the survival rate of all-ceramic crowns made from zircon ceramic after 1.2 million load cycles in a dual axis chewing simulator;
3. to measure the load to fracture of all-ceramic crowns made from zircon ceramic after 1.2 million load cycles in a dual axis chewing simulator.
2. Materials and methods
To test the new ZrSiO4 ceramic (Everest HPC, KaVo, Biberach, Germany) regarding its suitability for dental restorations, a number of tests were carried out. This included flexural strength testing, calculation of the Weibull modulus, the measurement of the fracture toughness, as well as the chemical solubility.
2.1. Material tests
2.1.1. Flexural strength
Flexural strength was measured using a biaxial piston-on-3-ball test. Twelve disk shaped specimens were prepared with a diameter of 15.1 ± 0.1 mm and a thickness of 1.2 ± 0.1 mm. The specimens were measured with a micrometer screw and their actual measures recorded. Then, the specimens were placed in the testing device and testing was then carried out in a universal testing machine (Zwick, Germany) at a crosshead speed of 1.5 mm/min. The flexural strength σ was calculated using the formula:
where ; υ = 0.19 (Poisson's parameter for ceramic); r1 = 5.97 mm (radius of the circle of the supporting balls); r2 = 0.76 mm (radius of the piston); r3 = radius of the specimen in mm; Fmax = fracture load in N; d = thickness of the specimen in mm.
In addition, the Weibull strength σ0 and the Weibull modulus m were calculated using linear regression analysis. The fracture probability Pσ is defined by the following equation: Pσ = (n − 0.5)/N, where N is the total number of specimen and n refers to the ascending order of the σ-values. The natural logarithms of σ are calculated and entered as x-values. The corresponding y-values are calculated using the formula ln (ln (1/(1 − Pσ))) and are then inserted into the coordinate plane. The diagram shows a straight line with the slope m and the intersection with the x-axis ln σ0.
2.1.2. Fracture toughness
Fracture toughness was measured using 10 box shaped specimens measuring 20 mm × 3 mm × 1.5 mm. Before testing, all measures of the specimens were recorded. Then, a notch was prepared within each specimen to a depth of 50% using a diamond jig saw. The width of the notch was defined by the thickness of the saw blade of approximately 300 μm. The specimens were then fractured in the universal testing machine at a crosshead speed of 1.5 mm/min. The exact depth of the slot was measured under a stereo microscope (Zeiss, Germany) using image analysis software (Soft-Imaging Software GmbH, Germany). The fracture toughness KIc was then calculated using the formula:
where Γm=1.9887−1.326a′−((3.49−0.68a′+1.35a′2)a′(1−a′))/(1+a′)2, Fmax = fracture load in N; L1, L2 = support distance; a′ = relative slot length a/W; B = specimen width in mm; W = specimen height in mm.
2.1.3. Chemical solubility
For the chemical solubility tests, 20 disc shaped specimens with a thickness of 1.2 ± 0.1 mm and a diameter of 15.0 ± 0.1 mm were used. The discs were cleaned in demineralized water and dried in an incubator for 4 h at 150 °C and then weighed on a precision scale to 0.1 mg accuracy. Then, the specimens were immersed in 4% acetic acid for 16 h at 80 °C. The cleaning and drying process was repeated and the specimens were weighed again. The initial and the final weights were used to calculate the chemical solubility using the formula:
where m0 = initial specimen weight; mn = weight after immersion in acetic acid; A = specimen surface (2πr2 + 2πrh).
2.2. Testing of CAD/CAM machined crowns
Sixty-four caries-free human molars were selected for similar size, cleaned by scaling and stored in 0.1% thymol solution throughout the course of the study. All teeth were visually examined and only those that were found to be sound and free from defects and cracks were used. Teeth were selected for the study if the variation in length and width was within 1 mm compared to the mean values (length 20 mm; diameter 10.5 mm). They were randomly assigned into 2 groups of 16 samples and 1 test group of 32 samples.
A representative clinical model with a missing first molar was selected  and . The selected model was embedded using a silicone putty material (Formasil Xact, Heraeus Kulzer, Germany). The occlusal table of the model was set parallel to the horizontal plane. Afterwards, a silicone mold of this representative model was fabricated with a silicone impression material (Formasil Xact). This index was then used as the negative form for fixing all abutment teeth in the sample holder and as an aid for the wax-up of the final crowns.
It has been demonstrated that abutment mobility can be a decisive factor in the evaluation of fracture strength of ceramic materials . In order to imitate physiological tooth mobility, all roots of the abutment teeth were covered with an artificial periodontal membrane made out of gum resin (Anti-Rutsch-Lack, Welenko, Germany)  and . In order to conform to the biological width, each tooth was coronally covered with wax 2 mm short of the cemento-enamel junction and then dipped once in the gum resin. After the gum resin had dried, the excess resin on the root tip was removed using a scalpel so that a uniform coating remained on the root surface.
The natural teeth were fixed into the silicone mold, the sample holder was attached to the prepared silicone mold and self-curing polyester resin (Technovit 4000, Heraeus Kulzer, Germany) was mixed and poured into the sample holder. After the resin had set, the silicone mold was removed and the crowns of the abutment teeth were cleaned.
An occlusal reduction of 1.8–2.0 mm was prepared with diamond burs leaving an abutment height of approximately 4.5 mm, followed by a circular 1.2 mm wide shoulder. Teeth were prepared with a 6° convergence angle.
Light body silicon impression material (Monopren®, Kettenbach, Germany) was syringed around the abutments, and putty material (Monopren®) was used on the tray. After removing the impression from the abutment teeth, the master model was poured in type 4 dental stone (Everest Rock, KaVo, Germany). After a setting time of at least 45 min, the model was removed from the impression. Sectioned dies were prepared and two layers of dye spacer (Purargent) were applied.
Thirty-two test crowns were fabricated from Everest HPC ZrSiO4 ceramic. For this purpose, a full wax-up was fabricated for each crown using the silicone index as template. Then, the abutment teeth were scanned and a second scan was performed of the outer surface of the waxed crowns (KaVo Everest Scan, KaVo, Germany). The Everest CAD/CAM system was used to generate the data for the manufacturing of the crowns. Everest HPC blanks were mounted in the Everest Engine CAM milling unit and the crowns were milled in the green stage. Subsequently, a tempering process at a maximal temperature of 1575 °C and sintering time of up to 4 h was performed in the Everest therm unit to sinter the green bodies. During this reaction sintering process, first PMSS is pyrolyzed, then ZrSi2 is oxidized, and finally both ZrSiO4 is formed and the ceramic is sintered to dense bodies.
The finalized crowns were tried on the plaster dies and adjusted as required.
2.2.2. Porcelain-fused-to-metal crowns
The copings of all porcelain-fused-to-metal crowns crowns were waxed-up with a minimum thickness of 0.4–0.5 mm. After investing (GC-Vest) and burnout of the wax patterns all crowns were cast in a high precious gold alloy (V-Classic). After divesting, the frameworks were sandblasted and cleaned with jet steam, then heated in a furnace at 950 °C for 10 min to create an oxidation layer.
Vita Omega porcelain was used for veneering. The veneer thickness varied between 1.5 mm on the occlusal surface and 0.7–1 mm on the circumference.
2.2.3. IPS Empress 2 crowns
The copings of all Empress 2 control crowns (Ivoclar, Schaan, Switzerland) were fabricated with a wall thickness of at least 0.8 mm following the manufacturer's recommendations. The patterns were sprued and invested (IPS Empress investment). The automatic heat pressing process was performed in the EP 500 press furnace using standard parameters. After completion of the pressing and cooling cycle, rough divestment was carried out. Subsequently, the pressed copings were cleaned and the final tooth shape was built up using the IPS Empress 2 ceramic material and final staining and glazing were performed.
2.2.4. Luting procedure
Glass–ionomer cement (Ketac-Cem, 3M Espe, Germany) was used for the cementation of the ZrSiO4 and PFM restorations in this study. The inner surfaces of all crowns were air particle abraded with 100 μm aluminium oxide. The abutment teeth were cleaned with water spray for 30 s and dried with air for 60 s. Glass–ionomer cement (Ketac-Cem) was mixed and applied to the inner surfaces of all crowns using a small brush. The crowns were seated on the abutment teeth and held in place with finger pressure. Any excess cement was removed by wiping with foam pellets.
Variolink®II (Ivoclar) dual-curing resin cement was used for the cementation of the Empress 2 restorations in this study. The inner surfaces of Empress 2 crowns were etched for 20 s with IPS Empress Ceramic etch gel, rinsed with water for 30 s and dried. A silane coupling agent was applied with a brush and after 60 s dried with air. All abutments were then etched with 37% phosphoric acid for 30–60 s. They were cleaned with water spray and dried with air. Syntac Primer (Ivoclar) was applied and dispersed after 15–20 s with an air syringe. Syntac adhesive (Ivoclar) was applied and dispersed with air after 10 s. Heliobond (Ivoclar) was applied and dispersed with air. At the same time Variolink II was mixed and applied evenly on the inner surfaces of all retainers; then the crowns were seated on the abutments and held in place with finger pressure. All surfaces of the Empress 2 crowns were light-cured for 1 min.
2.2.5. Fatigue tests: thermomechanical loading
All specimens in each group were exposed to 1.2 million cycles of thermo-mechanical fatigue in a computer-controlled chewing simulator simulating clinically relevant conditions (Willytec, Germany) . A load of 49 N was applied in the center of the occlusal surface of the crowns using 6 mm diameter ceramic antagonist ball (Steatit®, Ceramtec, Germany)  and . The vertical movement of the antagonist was set at 6 mm and the horizontal way at 0.5 mm, which resulted in a cycle frequency of 1.3 Hz. During testing, all samples were allowed to reach a thermal equilibrium between 5 and 55 °C for 60 s each with an intermediate pause of 12 s, maintained by a thermostatically controlled liquid circulator.
During the dynamic loading, all samples were examined twice a day. Fractures of the tooth or any part of the restoration were recorded as a failure.
All specimens were thereafter loaded until fracture occurred using a universal testing machine (Zwick Z010/TN2S, Zwick, Germany). A 1 mm thick tin foil was placed over the occlusal surface to achieve a homogeneous stress distribution. A perpendicular load was applied to the occlusal surface of the molar at a crosshead speed of 2 mm/min. The loads required to fracture the samples were recorded with the Zwick testXpert® V 7.1 software.
The statistical analysis of the fracture load tests was performed using the Kruskal–Wallis test for the comparison of multiple groups and pair-wise comparisons with the Wilcoxon test (SPSS, Germany) at a significance level 0.05.
3.1. Material tests
3.1.1. Flexural strength
Flexural strength was measured using a biaxial bend setup. The mean diameter of the disk specimens (n = 12) was 15.1 mm and the average thickness was 1.2 mm. Based on the average fracture load of 281.7 N, a flexural strength of 328.3 MPa was calculated. Using the described approach, a Weibull strength of 338.6 MPa and a Weibull modulus m of 16.8 were found (Table 1).
Flexural strength and Weibull parameters of the novel ZrSiO4ceramic