To characterize the stress induced deformation of bi-axial flexure strength (BFS) test specimens during processing to provide an insight into sintering effects and associated BFS determination.
40 Vitadur-Alpha and 80 IPS e.max Ceram disc-shaped specimens were condensed and a sintered on a silicon nitride refractory tray under controlled firing and cooling parameters. The mean of the maximum deflection (μm) and Ra values (μm) were determined using a high resolution profilometer and were related to the orientation of the measured surface within the furnace. BFS testing of the subsequent groups ( n = 20) was performed and the data related to the measured deformation of the sintered specimens. A two-way analysis of variance (ANOVA) where factors were identified as surface state and firing orientation with post hoc Tukey’s tests was complemented by pair-wise comparisons with a Student’s t -test for each measurement ( P < 0.05).
The mean of the maximum deflection values and the mean BFS for Vitadur Alpha discs were not significantly influenced by firing orientation ( P = 0.248 and P = 0.284, respectively). However, the Ra values were significantly different ( P < 0.001). The two-way ANOVA revealed a significant impact on the mean of the maximum deflection measurements for surface state ( P < 0.001) and firing orientation ( P < 0.001) during sintering ( P < 0.001). The mean Ra values were not significantly influenced. The BFS of sintered IPS e.max Ceram discs was sensitive to firing orientation ( P < 0.001).
Conventional glass theory explains that residual thermal stress gradients induced during sintering can cause test specimen deformation which can alter the expected BFS data. The study demonstrates that variability such as firing orientation during sintering which is very rarely reported in the literature can have a significant impact on the reported BFS data and can confound its interpretation.
Clinical performance predictions for dentin bonded crown (DBC) restorations has largely been performed using load to failure testing of geometrically representative crowns or by assessing the strength of the ceramic materials used to produce the restoration . Load to failure testing in the ‘crunch the crown’ testing methodology had previously been advocated , but the test has essentially been made redundant from the materials scientists’ testing armamentarium . The main problem with the ‘crunch the crown’ test is that failure is initiated by compressive stress on the external surface of the crown rather than from the extension of defects resident on the internal ‘fit’ surfaces of the restoration under tensile loading as demonstrated by axisymmetric finite element analysis and quantitative fractography on failed restorations. Additional problems with the ‘crunch the crown’ test are exacerbated by difficulties in loading the crowns reproducibly due to anatomical form constraints and the test quantifies the load to failure rather than strength . Therefore researchers have focused on determining the tensile strength of DBC materials to provide for clinical failure predictions .
The measurement of the tensile strength of brittle materials under uni-axial flexure conditions has been determined as early as 1959 using parallel rectangular specimens subjected to three- or four-point flexural loading . However, the stress in the loaded section of the specimen is not uniform varying from zero in the neutral plane to a maximum at the outer surfaces which accentuates the effect of surface condition on the measured strength and test results are in excess of the true tensile strength . An alternative is the bi-axial flexure strength (BFS) test where a disc-shaped specimen is centrally loaded so that the maximum tensile stresses occur at the center and decrease rapidly with increased radial distance from the center of the disc . The technique eliminates edge failures routinely associated with testing under uni-axial flexural , offers controllable specimen geometry and simple sample preparation . Discs also have a surface to volume ratio more closely related to actual DBCs than uni-axial flexure specimens . Several variations of the BFS testing methodology exist but the ball-on-ring test was advocated by de With and Wagemans as the most reliable testing configuration.
With recent advances in ceramic technology, considerable differences of opinion exist among dental materials scientists and dental practitioners as to which dental ceramic system offers increased clinical performance for DBCs. Strength variations of ±25% of the mean are routine when laboratory testing dental ceramics in the ball-on-ring BFS configuration. Piddock et al. postulated that technical skill influenced the BFS and this was confirmed by the variously reported mean BFS values in the dental literature (134 ± 15 MPa and 179 ± 27 MPa ), for an ‘as-fired’ aluminous core porcelain (Vitadur-N, Vita Zahnfabrik, Bad Säckingen, Germany) tested under similar laboratory conditions. However, pre-measuring the relative proportions of powder and modeling liquid improved the reproducibility of the BFS data for the ‘as-fired’ Vitadur-N aluminous core porcelain (139 ± 19 MPa , 139 ± 10 MPa and 142 ± 13 MPa ) when using standardized condensation techniques. However, for dental ceramics manufactured by a sintered frit route the firing regime may introduce transient and residual stresses across the thickness of ceramic test specimens during processing . The relative position of the specimen within the furnace during sintering and its contact relationship with a supporting refractory tray has the potential to induce specimen variation which has not previously been considered in the dental literature. The aim of the current study was to characterize the stress induced deformation of BFS test specimens during processing to provide an insight into sintering effects and associated BFS determination.
Materials and methods
Specimen condensation and sintering regimen
Forty Vitadur Alpha dentin (Vita Zahnfabrik, Bad Säckingen, Germany) disc-shaped specimens were condensed into a perspex mold clamped to a glass-slide. A slurry consistency containing 0.6 g of Vitadur Alpha dentin powder (shade D2, Lot 1064) and 0.16 mL of Vita Modeling Fluid (Lot 14209R) was mixed and condensed under ultrasonic vibration (CeramoSonic TA II, Shofu, Kyoto, Japan). The water expelled under vibration was removed with absorbent paper. The specimen was leveled using a razor blade and the leveled surface was termed the challenged surface for the purpose of the current study as this surface had been modified during specimen manufacture. The discs were removed from the mold and placed onto a silicon nitride refractory tray with the unchallenged surface, which was in contact with the glass-slide during condensation and therefore had not been modified during specimen manufacture, was placed in contact with the refractory tray. The specimens were sintered in a vacuum furnace (Vita Vacumat 40, Vita Zahnfabrik, Bad Säckingen, Germany) where they were pre-dried for 6 min at 600 °C, heated under vacuum at 60 °C/min to 960 °C, the vacuum released and the specimens held for 1 min at 960 °C, before being air cooled to 60 °C at 2.9 °C/min. The unchallenged surface which was in contact with the glass-slide during condensation and the refractory tray during sintering was marked. The 40 specimens (13.06 ± 0.08 mm diameter and 1.06 ± 0.03 mm thickness—measured with a digital caliper and digital micrometer, respectively accurate to 10 μm (Mitutoyo Corp., Tokyo, Japan)) were randomly divided into two groups of 20 (Groups A and B) and stored in a desiccator prior to analysis.
IPS e.max ® Ceram dentin (Ivoclar-Vivadent AG, Schaan, Liechtenstein) discs were condensed into the same perspex mold assembly using the ultrasonic vibration technique and leveling procedure described above. A slurry consistency of 0.6 g of IPS e.max ® Ceram dentin (shade A2, Lot K49034) and 0.14 mL IPS e.max ® Ceram build-up liquid (Lot K49035) was used . The specimens were placed onto a silicon nitride refractory tray and sintered which involved pre-drying for 8 min at 403 °C, heating to 769 °C at 50 °C/min under vacuum, the vacuum released and the specimens held at 770 °C for 1 min before being air cooled to 60 °C at 2.9 °C/min. Forty IPS e.max ® Ceram specimens were sintered with the unchallenged surface placed in contact with the refractory tray. The unchallenged surface was marked and the discs (12.90 ± 0.15 mm diameter and 1.04 ± 0.03 thickness) were randomly allocated to two groups of 20 specimens (Groups C and D) prior to storage in a desiccator . An additional 40 IPS e.max ® Ceram discs were prepared and sintered with the challenged surface (the leveled surface during specimen condensation) placed in contact with the refractory tray. The unchallenged surface was marked and the specimens (12.88 ± 0.09 mm diameter and 1.02 ± 0.04 mm thickness) were randomly divided into two further groups of 20 specimens (Groups E and F) prior to storage in a desiccator .
Profilometric deflection test
The disc-shaped specimens were placed, aligned and held for the entire duration of the test in a custom-made metallic leveling device. A contact diamond stylus profilometer (Talysurf CLI 2000 Taylor-Hobson Precision, Leicester, UK) with a 90° conisphere stylus tip of 2 μm radius was used to characterize the mean of the maximum deflection of the unchallenged surface of the Vitadur Alpha (Groups A and B) and IPS e.max ® Ceram (Groups C–F) dentin discs. Profilometric measurements were performed at a stylus velocity of 1 mm/s with an applied force of 0.75 mN, across two perpendicular 5 mm 2 areas (10 mm length and 0.5 mm width) coincident with the center of the specimen. Each 5 mm 2 area was composed of 126 traces with a 4 μm step-size ( y -direction) with data points recorded every 10 μm ( x -direction) to a 40 nm resolution ( z -direction). The mean of the maximum deflection (μm) of the unchallenged surface of the dentin discs was determined by averaging the characterized perpendicular mean of the maximum deflection values obtained from the 5 mm 2 areas .
Additionally the surface roughness of the unchallenged and challenged surfaces of the Vitadur Alpha (Groups A) specimens and the IPS e.max ® Ceram (Groups C and E) dentin specimens was determined. The ISO recognized roughness parameter (Ra value) which represents ‘the arithmic mean of the absolute departures of the roughness profile from the mean line’ was estimated using software (Talymap 2.6.0, Taylor-Hobson Precision, Leicester, UK). The estimated Ra value was quantified from the profilometric profiles generated across the two perpendicular 5 mm 2 areas (composed of 126 traces) at the operating conditions of stylus velocity, applied force and step-size outlined above. All specimens from each group investigated (Groups A, C and E) were randomly selected and the mean Ra value was determined using a 2.5 mm cut-off Gaussian filter.
Bi-axial flexure strength test
Following characterization of the mean of the maximum deflections (μm) and the Ra values, the BFS of the as-fired disc-shaped dentin specimens were determined using a ball-on-ring test configuration . The discs were supported by a 10 mm diameter metallic ring and centrally loaded with a 2.75 mm diameter spherical ball indenter at a cross-head speed of 1.0 mm/min. A thin rubber film was placed between the specimen and the metallic ring to uniformly distribute the load. The load at fracture was recorded and the BFS was calculated using Eq. (1) .
where σ max is the maximum tensile stress, P the measured load to fracture, a the radius of the knife-edged support, ν is Poisson’s ratio for the material (values of 0.25 and 0.23 are used for Vitadur Alpha and IPS e.max ® Ceram , respectively) and h is the specimen thickness.
Twenty Vitadur Alpha dentin discs (Group A) were loaded with the unchallenged surface in compression, namely in contact with the spherical ball indenter. The remaining 20 discs (Group B) were loaded with the unchallenged surface in tension, namely in contact with the metallic ring. Similarly, for the IPS e.max ® Ceram discs, 20 specimens (Group C) were loaded with the unchallenged surface placed in compression and the remaining 20 discs (Group D) were loaded with the unchallenged surface placed in tension. The IPS e.max ® Ceram discs sintered with the challenged surface in contact with the refractory tray (the unchallenged surface in contact with air) were loaded with the unchallenged surface placed in compression (Group E) and tension (Group F).
The statistical significance of the characterized mean of the maximum deflection measurements on the unchallenged surface of the Vitadur Alpha discs (Groups A and B) was determined using a Student’s t -test at a significance level of P < 0.05. Similarly, the statistical significance of the characterized mean of the maximum deflection measurements on the unchallenged surface of the IPS e.max ® Ceram discs in contact with the refractory tray (Groups C and D) or air (Groups E and F) during sintering were determined using a Student’s t -test. The statistical significance of the mean Ra value on the challenged and unchallenged surface of the Vitadur Alpha (Group A) were determined using a Student’s t -test ( P < 0.05) and IPS e.max ® Ceram discs (Groups C and E) were determined using a two-way analysis of variance (ANOVA) where factors were identified as surface (challenged and unchallenged) and firing orientation (contact with the refractory tray or air) with post hoc Tukey’s tests ( P < 0.05) (SPSS version 16, Inc., Chicago, IL, USA).
The statistical significance of the mean BFS of the Vitadur Alpha discs tested with the unchallenged surface tested in compression and tension (Groups A and B, respectively) was determined using a Student’s t -test at a significance level of P < 0.05. Additionally, the statistical significance of the mean BFS of the IPS e.max ® Ceram discs tested with the unchallenged surfaces in compression and tension, when sintered in contact with the refractory tray (Groups C and D, respectively) or air (Groups E and F, respectively), were determined using a two-way ANOVA and post hoc Tukey’s tests at P < 0.05 (SPSS version 16, Inc., Chicago, IL, USA).